Cell Metabolism

Review

AMPK and TOR: The Yin and Yang of Cellular

Nutrient Sensing and Growth Control
Asier González,1 Michael N. Hall,1 Sheng-Cai Lin,2 and D. Grahame Hardie3,*
1Biozentrum, University of Basel, CH4056 Basel, Switzerland
2School of Life Sciences, Xiamen University, Xiamen, 361102 Fujian, China
3Division of Cell Signalling & Immunology, School of Life Sciences, University of Dundee, Dundee, DD1 5EH, Scotland, UK

*Correspondence: d.g.hardie@dundee.ac.uk
https://doi.org/10.1016/j.cmet.2020.01.015

The AMPK (AMP-activated protein kinase) and TOR (target-of-rapamycin) pathways are interlinked,
opposing signaling pathways involved in sensing availability of nutrients and energy and regulation of cell
growth. AMPK (Yin, or the ‘‘dark side’’) is switched on by lack of energy or nutrients and inhibits cell growth,
while TOR (Yang, or the ‘‘bright side’’) is switched on by nutrient availability and promotes cell growth. Genes
encoding the AMPK and TOR complexes are found in almost all eukaryotes, suggesting that these pathways
arose very early during eukaryotic evolution. During the development of multicellularity, an additional tier of
cell-extrinsic growth control arose that is mediated by growth factors, but these often act by modulating
nutrient uptake so that AMPK and TOR remain the underlying regulators of cellular growth control. In this re-
view, we discuss the evolution, structure, and regulation of the AMPK and TOR pathways and the complex
mechanisms by which they interact.

All eukaryotic cells are now thought to have arisen via a single
endosymbiotic event when an archaeal host cell engulfed bacte-
ria that were capable of oxidative metabolism, the latter eventu-
ally becoming mitochondria (Lane, 2006; Sagan, 1967). This
event was followed by the transfer of most of the genes from
the genome of the endosymbiont to that of the host—it has
been argued that this separation of energy-generating capacity
from gene expression allowed a large increase in the energy
available per gene, thus permitting a major expansion in gene
number in the host (Lane and Martin, 2010). This may in turn
have enabled major enhancements in the complexity of eukary-
otic cells compared with their prokaryotic counterparts,
including the development of endomembrane systems such as
lysosomes or vacuoles (de Duve, 2005), and the associated traf-
ficking of materials between these internal compartments and
the plasma membrane via membrane-bound vesicles. New
cellular functions this led to were phagocytosis and pinocytosis,
used by many protists today as mechanisms of feeding, and
autophagy, used by all eukaryotic cells for recycling of cellular
components that are damaged or surplus to requirements, or
as an emergency measure during nutrient starvation. Phagocy-
tosis, pinocytosis, and autophagy deliver proteins, lipids, and
carbohydrates, or even whole organelles, such as mitochondria,
to lysosomes or vacuoles; the latter are acidic compartments,
where the engulfed materials are broken down to recycle their
components either for catabolism or re-use. Lysosomes or vac-
uoles can therefore be considered to be the ‘‘gut’’ or digestive
systems of unicellular eukaryotes, particularly in amoeboid pro-
tists that feed by phagocytosis or pinocytosis. They would there-
fore have been a major source of nutrients and appear to have
developed into hubs for nutrient sensing, as discussed below.
As these processes were evolving, early eukaryotes would
have needed signaling pathways that could monitor the function
of their new internal organelles and regulate cell growth and pro-
liferation accordingly. For example, there would have been a
need to monitor the output of ATP by mitochondria and to upre-
gulate their ATP-generating capacity if or when the supply of ATP
was insufficient; this is now a major function of the AMPK (AMP-
activated protein kinase) signaling pathway. In addition, there
would have been a requirement to monitor the supply of
nutrients, such as amino acids and glucose, produced at the
lysosome by phagocytosis, pinocytosis, or autophagy and to up-
regulate cell growth when these nutrients were available; this is
now a key function of the TOR (target-of-rapamycin) pathway.
We propose that these two opposing pathways, which are pre-
sent in almost all present-day eukaryotes, are the descendants
of ancient nutrient sensing and signaling pathways that arose
very early during eukaryotic evolution. AMPK represents the
Yin (‘‘dark’’ or ‘‘passive’’) side that signals lack of nutrients or
insufficient ATP and inhibits cell growth, whereas TOR repre-
sents the Yang (‘‘bright’’ or ‘‘active’’) side that signals availability
of nutrients and promotes cell growth. Just as in the Chinese phi-
losophy of Taoism from which the Yin-Yang concept is derived,
an appropriate balance between these two opposing elements
ensures homeostasis and thus a healthy cell or organism.
In present-day unicellular eukaryotes, including fungi like
Saccharomyces cerevisiae, growth and proliferation are regu-
lated almost entirely by nutrient availability, and the orthologs
of AMPK and TOR play crucial roles in this. However, during
the development of multicellular organisms, the uptake (and
hence the intracellular availability) of nutrients has become
modulated by an additional tier of cell-extrinsic regulation medi-
ated by growth factors and cytokines (Palm and Thompson,
2017). It can be argued that these cell-extrinsic factors ‘‘license’’
or allow cells to take up nutrients but that the AMPK and TOR
pathways, which sense intracellular nutrient availability, remain
the primary internal regulators of cell growth and proliferation.
Interestingly, most of the mutations that cause cancer in multi-
cellular organisms appear to affect the higher-level, cell-extrinsic
regulation of cell growth. Such mutations allow cancer cells to

472 Cell Metabolism 31, March 3, 2020 ª 2020 Elsevier Inc.

Cell Metabolism

Review

Table 1. Gene and Protein Names for AMPK, TORC1, and TORC2 Subunit Orthologs in Humans and Different Model Organisms
(Arabidopsis thaliana, Drosophila melanogaster, Caenorhabditis elegans, Saccharomyces cerevisiae, Schizosaccharomyces pombe)
Signaling H. sapiens A. thaliana D. melanogaster C. elegans S. cerevisiae S. pombe
complex Gene Protein Locus Protein Gene Protein Gene Protein Gene Protein Gene Protein
AMPK PRKAA1 a1 AT3G01090 KIN10 snfA AMPKa aak-1 AAK-1 SNF1 Snf1 ssp2 Ssp2
PRKAA2 a2 AT3G29160 KIN11 aak-2 AAK-2 ppk9 Ppk9
PRKAB1 b1 AT5G21170 KINb1 alc AMPKb aakb-1 AAKB-1 SIP1 Sip1 amk2 Amk2
PRKAB2 b2 AT4G16360 KINb2 aakb-2 AAKB-2 SIP2 Sip2
AT2G28060 KINb3a GAL83 Gal83
PRKAG1 g1 AT3G48530 KINgb SNF4Ag AMPKg aakg-1 AAKG-1 SNF4 Snf4 cbs2 Cbs2
PRKAG2 g2 AT1G09020 KINbgc aakg-2 AAKG-2
PRKAG3 g3 aakg-3 AAKG-3
aakg-4 AAKG-4
aakg-5 AAKG-5
TORC1 MTOR mTOR AT1G50030 TOR tor TOR let-363 TOR TOR1/TOR2 TOR1/TOR2 tor2 Tor2
MLST8 mLST8 AT3G18140 LST8-1 lst8 LST8 mlst-8 LST-8 LST8 Lst8 pop3 Pop3
AT2G22040 LST8-2d
RPTOR RAPTOR AT5G01770 RAPTOR1A raptor RAPTOR daf-15 DAF-15 KOG1 Kog1 mip1 Mip1
AT3G08850 RAPTOR1B
TORC2 MTOR mTOR AT1G50030 TOR tor TOR let-363 TOR TOR2 TOR2 tor1 Tor1
MLST8 mLST8 AT3G18140 LST8-1 lst8 LST8 mlst-8 LST-8 LST8 Lst8 pop3 Pop3
AT2G2204 LST8-2d
RICTOR RICTOR - - rictor RICTOR rict-1 RICTOR AVO3 Avo3 ste20 Ste20
MSIN1 mSIN1 - - sin1 SIN1 sinh-1 SIN1 AVO1 Avo1 sin1 Sin1
a
This is an unusual plant-specific b subunit that contains the C-terminal domain but lacks a CBM (carbohydrate-binding module).
b
By sequence, this appears to be an ortholog of mammalian g subunits, but it does not appear to form functional heterotrimers (Zhao, 2019).
c
This is an unusual plant-specific g subunit that contains a CBM (carbohydrate-binding module) fused to the four CBS motifs found in other AMPK-g
subunits.
d
This is probably a non-functional protein (Moreau et al., 2012).

become ‘‘rebels’’ that have partially reverted to their unicellular
origins and that switch over to using cell-intrinsic growth control,
based on nutrient availability and controlled by the AMPK and
TOR pathways.
Yin: The Structure and Regulation of AMPK/SNF1
Complexes
Subunit Structure and Evolution
AMPK appears to occur universally as heterotrimeric complexes
comprising catalytic a ubunits and regulatory b and g subunits
(Ross et al., 2016b). Genes encoding all three subunits are
readily found within the genomes of almost all eukaryotes
(Table 1 and Figure 1). However, the orthologs in budding yeast
(S. cerevisiae) and plants are not allosterically activated by AMP
and were discovered independently of mammalian AMPK by ge-
netic approaches (Alderson et al., 1991; Celenza and Carlson,
1986). They are therefore not usually referred to as AMPK but
instead in yeast as Snf1 complexes (SNF1 being the gene encod-
ing the catalytic subunit) and in plants as Snf1-related kinase-1
(SnRK1) complexes.
Interestingly, the only eukaryotes known to lack AMPK subunit
orthologs are parasites that spend all or most of their life cycle
living inside other eukaryotic cells, including Encephalitozoon
cuniculi and Plasmodium falciparum, the latter being the causa-
tive agent of human malaria (Figure 1). These parasitic eukary-
otes appear to have undergone stringent selection for small
genome size, with E. cuniculi having one of the smallest known
genomes of any eukaryote, encoding only 29 conventional and
3 atypical protein kinases (compared with >500 in humans)
(Miranda-Saavedra et al., 2007). Ancestors of these organisms
most likely did have AMPK genes, but the modern-day descen-
dants may have been able to dispense with them because the
host cell would provide AMPK that regulates cellular energy bal-
ance on their behalf. Consistent with this, species closely related
to P. falciparum that cause malaria in birds (P. gallinaceum and
P. relictum) do still have conventional AMPK genes (Böhme
et al., 2018). Interestingly, TOR genes are missing in E. cuniculi
and P. falciparum (Figure 1) but are also absent in
P. gallinaceum and P. relictum.
Mammals, including humans, have two genes encoding iso-
forms of AMPK-a (a1 and a2), two encoding AMPK-b (b1 and
b2), and three encoding AMPK-g (g1, g2, and g3) (Table 1).
These multiple isoforms appear to have arisen during the
two rounds of whole-genome duplication that occurred during
the early evolution of vertebrates (Ross et al., 2016b). All
twelve combinations of these subunit isoforms are able to
form heterotrimeric complexes, although it is not certain that
all combinations exist in vivo. Structures for several almost-
complete human AMPK heterotrimers, i.e., a2b1g1 (Xiao
et al., 2013), a1b1g1 (Calabrese et al., 2014), a1b2g1 (Li
et al., 2015), and a2b2g1 (Ngoei et al., 2018), have been ob-
tained via X-ray crystallography. The complexes were all crys-
tallized in active conformations, and their structures are
very similar; a schematic representation of a generalized
AMPK heterotrimer based on these structures is shown in
Figure 2.

Cell Metabolism 31, March 3, 2020 473


Structure of AMPK and Canonical Adenine Nucleotide
(Energy)-Sensing Mechanism
Although the main theme of this review is nutrient sensing, we will
first discuss the classical or ‘‘canonical’’ mechanism by which
AMPK responds to the changing energy status of cells. The cata-
lytic a subunits of AMPK contain, at their N-termini, conventional
serine/threonine kinase domains with a small N-lobe and larger C-
lobe and the catalytic site in the cleft between them. As with many
other members of the ePK (eukaryotic protein kinase) family,
AMPK complexes are only significantly active when phosphory-
lated at a critical residue within the activation loop, a stretch of
z20 amino acids in the C-lobe between the highly conserved
DFG and APE motifs. In AMPK, the critical phosphorylation site
is a threonine, usually referred to as Thr172 after its position in
the rat a2 sequence where originally mapped (Hawley et al.,
1996). Thr172 is not phosphorylated by AMPK itself but by up-
stream kinases, principally by LKB1 (liver kinase B1) (Hawley
et al., 2003; Shaw et al., 2004; Woods et al., 2003), the active
form of which is a heterotrimeric complex also containing
STRAD-a or –b, and the scaffold protein MO25-a or –b (Zeqiraj
et al., 2009). LKB1 was originally identified as the product of the
tumor suppressor gene STK11, which is mutated in Peutz-Jegh-
ers Syndrome (an inherited susceptibility to cancer) as well as in
some sporadic (i.e., non-inherited) cancers, especially lung ade-
nocarcinomas (Alessi et al., 2006; Ji et al., 2007; Sanchez-Ces-
pedes et al., 2002). Although LKB1 was subsequently shown to
phosphorylate and activate twelve other kinases with kinase do-
mains related to AMPK (the AMPK-related kinase family; Jaleel
et al., 2005; Lizcano et al., 2004), AMPK was the first downstream
target for LKB1 to be identified, and this introduced an intriguing
connection between AMPK and cancer. Indeed, it is now clear
that AMPK can also act as a tumor suppressor, at least in certain
animal models of cancer (Vara-Ciruelos et al., 2019).
A summary of the canonical and non-canonical mechanisms
that activate AMPK, and selected downstream targets involved
474 Cell Metabolism 31, March 3, 2020
Cell Metabolism
Review
Figure 1. Conservation of TOR and AMPK
Signaling Components among Eukaryotic
Species
Black boxes indicate presence, and white boxes
absence, of the indicated genes/proteins in the
corresponding organisms (Tatebe and Shiozaki,
2017; van Dam et al., 2011). Gray boxes indicate
limited similarity to the human counterpart. There is
no evidence that the S. cerevisiae Rheb regu-
lates TORC1.
in its promotion of catabolic processes, in-
hibition of anabolic processes, and effects
on DNA replication, are shown in Figure 3.
In the canonical mechanism that is en-
shrined in its name, AMPK is activated by
binding of 50 -AMP, with activation occur-
ring not by one but three mechanisms: (1)
allosteric activation of AMPK already
phosphorylated on Thr172 (Carling et al.,
1987; Ferrer et al., 1985; Yeh et al., 1980);
(2) enhanced Thr172 phosphorylation by
the LKB1 complex (Hawley et al., 1995);
and (3) protection against Thr172 dephos-
phorylation by protein phosphatases (Davies et al., 1995). All
three effects are due to binding of AMP to AMPK, not to the up-
stream kinase or phosphatase, and this tripartite mechanism en-
sures that the system responds to small increases in AMP in a
very sensitive manner. Although there is general agreement
that only AMP binding causes effect #1 above, ADP binding simi-
larly triggers effects #2 and #3 (Oakhill et al., 2011; Xiao et al.,
2011). However, most AMPK complexes (apart from those con-
taining the g2 isoform) are about 10-fold more sensitive to AMP
than ADP, suggesting that increases in AMP are the primary acti-
vating signal, although increases in ADP may contribute (Ross
et al., 2016a). All of the activating effects of AMP and ADP are
antagonized by binding of ATP so that the AMPK system effec-
tively monitors cellular AMP:ATP and ADP:ATP ratios.
Where are the regulatory binding sites where these adenine
nucleotides are sensed? The g subunits contain four tandem re-
peats of a sequence termed a CBS (cystathionine b-synthase)
motif (Bateman, 1997). These occur, usually as just two tandem
repeats, in about 75 proteins in humans, and they are also found
in archaea and bacteria. Single pairs of tandem CBS repeats
associate into pseudodimers (termed Bateman modules), poten-
tially creating two pseudo-symmetrical ligand-binding sites in
the intervening cleft, although in many cases, only one is utilized.
These sites usually bind ligands containing adenosine or (less
often) guanosine (Anashkin et al., 2017; Scott et al., 2004). The
two Bateman modules in each AMPK-g subunit associate
head-to-head to form a flattened disk with four potential binding
sites for adenine nucleotides in the center (Figure 2). However,
only three are utilized, i.e., CBS3, which is accessible from one
face of the g subunit, and CBS1 and CBS4, accessible from
the other. The critical site appears to be CBS3; the a-linker, a
flexible region of the a subunit that connects the a-AID
(a-auto-inhibitory domain) and a-CTD (a-C-terminal domain),
wraps around the face of the g subunit containing CBS3, con-
tacting its bound AMP (Figure 2). This interaction is not thought
Cell Metabolism
Review
to occur when ATP is bound at CBS3 instead of AMP, and the
consequent release of the a-linker from the g subunit is pro-
posed to allow the a-AID to rotate back into its inhibitory position
behind the kinase domain (Chen et al., 2009; Chen et al., 2013; Li
et al., 2015; Xiao et al., 2011; Xin et al., 2013); this model thus ex-
plains allosteric activation by AMP as well as its antagonism by
ATP. At the same time, the resulting conformational changes
may alter the accessibility of Thr172 for phosphorylation and/or
dephosphorylation, although those aspects of the mechanism
are less well understood. The functions of the CBS1 and CBS4
sites are less clear, although they are close to the CBS3 site in
the center of the CBS repeats, where the three sites interact.
One proposal is that CBS1 binds ATP permanently, while
CBS4 binds AMP permanently, and that these constitutive bind-
ing events alter the conformation of the CBS3 site such that it has
a higher affinity for AMP than ADP or ATP (Gu et al., 2017b). This
helps to explain how AMPK achieves the difficult task of sensing
changes in AMP in the 30–300 mM range despite the presence of
mM concentrations of ATP (Gowans et al., 2013). An additional
explanation is that only Mg2+-free ATP competes with AMP at
the CBS3 site (Pelosse et al., 2019), although 90% of intracellular
ATP is thought to be present at the Mg.ATP2- complex. Accord-
ing to this model, the ATP and AMP constitutively bound at the
CBS1 and CBS4 sites, respectively, act essentially as regulatory
co-factors. This explains why a functional CBS4 site is required
for activation even when overall AMP levels remain at the basal
level (Zong et al., 2019).
Although the sequences of the a, b, and g subunits are well
conserved, the regulation by adenine nucleotides of AMPK ortho-
logs from eukaryotes other than mammals is much less well stud-
ied. As mentioned earlier, neither Snf1 complexes from
S. cerevisiae (Wilson et al., 1996) nor SnRK1 complexes from
plants (Mackintosh et al., 1992) appears to be allosterically acti-
vated by AMP, although the dephosphorylation of the threonine
Figure 2. Schematic View of the Structure of
the AMPK Heterotrimer
The diagram is a composite derived from the
structures of the human a2b1g1 (Xiao et al., 2013),
a1b1g1 (Calabrese et al., 2014), and a1b2g1 (Li
et al., 2015) complexes and is an active conforma-
tion with Thr172 phosphorylated and three mole-
cules of AMP bound to the g subunit. The a subunit
is shown in yellow (apart from the ST loop, in red),
the b subunit in lilac, and the g subunit in blue-green.
The a-linker is depicted as a yellow chain connect-
ing the a-AID and the a-CTD, and it contacts AMP
bound in the CBS3 site. The ST loop is not resolved
in any of the structures, and its exact positioning is
speculative. The N-terminal regions of the b sub-
units, and the linker between the b-CBM and the
b-CTD, (not shown) are either absent or are not
resolved in any of the structures.
residues equivalent to Thr172 were re-
ported to be inhibited by ADP in
S. cerevisiae (Mayer et al., 2011) and by
AMP in plants (Sugden et al., 1999a). Allo-
steric activation by AMP has been reported,
although not well studied, using the com-
plexes from D. melanogaster (Pan and Har-
die, 2002), C. elegans (Apfeld et al., 2004),
and S. pombe (Forte et al., 2019). It seems possible that allosteric
activation, which is physiologically significant in intact cells
(Gowans et al., 2013), was a later evolutionary refinement that
increased the overall sensitivity of the system to small changes
in AMP.
Non-canonical Activation of AMPK by Ligands Binding at
the ADaM Site
The heterotrimeric AMPK complex contains other ligand-binding
sites whose physiological function remains less clear. One is the
glycogen-binding site on the b-CBM (b-carbohydrate-binding
module), which is present in the b subunits of all eukaryotes
and in mammalian cells and causes a proportion of AMPK to
bind to glycogen (Hudson et al., 2003; Polekhina et al., 2003;
Polekhina et al., 2005). Intriguingly, as well as a conventional
CBM on the b subunit, many higher plant SnRK1 complexes
also contain a second CBM fused at the N terminus of the g sub-
unit, forming a so-called bg subunit (Lumbreras et al., 2001;
Zhao, 2019). Although it has been proposed that the single
CBM of mammalian AMPKs may allow them to sense the struc-
tural state of glycogen (McBride et al., 2009), more work is
required to confirm that hypothesis. Another ligand-binding
site lies in a cleft (termed the ADaM site) between the other
face of the CBM (i.e., opposite to the glycogen-binding site)
and the N-lobe of the kinase domain on the a subunit (Figure 2).
Several ligands that bind in this site cause a dramatic allosteric
activation of AMPK with, usually, a more modest effect to pro-
mote net Thr172 phosphorylation (Göransson et al., 2007;
Sanders et al., 2007; Scott et al., 2014; Yan et al., 2019). Howev-
er, a curious feature is that with the exception of salicylate (a nat-
ural product of plants, but not animals) (Hawley et al., 2012), all of
the compounds currently known to bind there are synthetic mol-
ecules that emerged from high-throughput screens searching for
allosteric activators of AMPK (e.g., Cokorinos et al., 2017; Cool
et al., 2006; Myers et al., 2017). This binding site is therefore a
Cell Metabolism 31, March 3, 2020 475
 Cell Metabolism

Review

Figure 3. Canonical and Non-canonical Mechanisms of AMPK Activation

Proteins shown in green promote activation of AMPK, while proteins shown in red promote inhibition (aldolase is a positive effector when unoccupied by FBP).
Canonical activation by energy stress requires LKB1, occurs in the cytoplasm, and is triggered by increases in AMP:ATP or ADP:ATP ratios. By contrast, non-
canonical activation by glucose starvation involves translocation of AXIN:LKB1 to the lysosome, where a pool of AMPK myristoylated on the b subunit resides
permanently and can occur in the absence of any changes in adenine nucleotides. Non-canonical activation by Ca2+ ions released from the ER or within the
nucleus, triggered by hormones or DNA damage, respectively, requires CaMKK2 and not LKB1. Note that the localized increase in Ca2+ caused by activation of
TRPV channels is not sufficient to activate CaMKK2. See main text for details.

type of ‘‘orphan receptor,’’ and many researchers in the field sus-
pect that there is a unidentified metabolite occurring in animal
cells that binds to it, hence the acronym ADaM (allosteric drug
and metabolite) site (Langendorf and Kemp, 2015).
Non-canonical Activation of AMPK by Ca2+ and by DNA
Damage
Thr172 can also be phosphorylated by alternate upstream ki-
nases, including the Ca2+/calmodulin-dependent kinase,
CaMKK2 (Hawley et al., 2005; Hurley et al., 2005; Woods et al.,
2005) and TAK1 (Transforming growth factor-b-Activated Ki-
nase-1) (Momcilovic et al., 2006). The physiological importance
of TAK1 as a means of AMPK activation is not well established,
although there is one report that it is involved in AMPK activation
in response to TRAIL (tumor necrosis factor-related apoptosis-
inducing ligand) (Herrero-Martı́n et al., 2009). By contrast, there
is good evidence that AMPK can be activated by the CaMKK2
pathway in response to hormones or growth factors that trigger
release of Ca2+ from the endoplasmic reticulum (Figure 3). This
includes hormones acting at G protein-coupled receptors linked
via Gq/G11 to release of the Ca2+-mobilizing messenger IP3
(inositol-3,4,5-trisphosphate), such as thrombin acting at prote-
ase-activated receptor-1 in endothelial cells (Stahmann et al.,
2006), acetylcholine acting at M3 muscarinic receptors in various
cell types (Jadeja et al., 2019; Merlin et al., 2010; Thornton et al.,
2008; Xue et al., 2016), and ghrelin acting at GHSR1 receptors in
neurons of the hypothalamus (Yang et al., 2011). AMPK is also
activated via a Ca2+/CaMKK2-dependent mechanism by the
growth factor VEGF (vascular endothelial growth factor) acting
at the tyrosine kinase-linked VEGF receptor in endothelial cells,
which triggers release of IP3 via activation of PLCg (phospholi-
pase C-g) (Reihill et al., 2007; Stahmann et al., 2010).
Another non-canonical AMPK activation mechanism occurs in
response to DNA damage and/or replicative stress (Figure 3),
which can be induced by etoposide, hydroxyurea, aphidicolin,
or ionizing radiation (Fu et al., 2008; Li et al., 2019b; Sanli
et al., 2010). Interestingly, the effects of etoposide, hydroxyurea,
or aphidicolin require CaMKK2 but not LKB1, correlate with in-
creases in nuclear Ca2+, only activate AMPK in the nucleus,
and (at least for etoposide) only activate the a1 isoform (Li
et al., 2019b; Vara-Ciruelos et al., 2018). Studies with AMPK

476 Cell Metabolism 31, March 3, 2020

Cell Metabolism
Review
knockout cells reveal that they are hypersensitive to cell death
induced by DNA damage or replicative stress (Vara-Ciruelos
et al., 2018), and this correlates with increased resection of repli-
cation forks as well as other chromosomal abnormalities (Li et al.,
2019b). The defects in the knockout cells have been attributed,
at least in part, to lack of phosphorylation by AMPK of the 50 -30
exonuclease EXO1, which normally causes its association with
14-3-3 proteins, thus restraining its ability to resect replication
forks (Li et al., 2019b). Since many of these genotoxic treatments
are used in cancer therapy, it seems likely that they would be
more efficacious if administered together with an AMPK inhibi-
tor, thus preventing the protective effects of AMPK against cell
death induced by DNA damage or replicative stress.
Non-canonical Activation of AMPK by Glucose
Starvation
Recent studies in mammalian cells have revealed, perhaps sur-
prisingly, that activation of AMPK in response to glucose starvation
can occur via a non-canonical, AMP-independent mechanism.
The first clues came from administration of siRNAs targeting
AXIN1 into the tail vein of mice, using adenoviral vectors that direct
expression to the liver. After overnight starvation, animals receiving
siRNA showed diminished AMPK activation and increased fat stor-
age in liver. This led to the discovery that AXIN1, which was initially
identified as a central scaffold protein for Wnt signaling (Zeng et al.,
1997), binds constitutively to LKB1 and acts as an adaptor for
LKB1 to associate with and phosphorylate AMPK; this initial char-
acterization of the role of AXIN1 was based on an in vitro reconsti-
tution experiment where high levels of AMP were required for the
interaction to occur (Zhang et al., 2013), which can now be classi-
fied as a cytosolic, AXIN/AMP-dependent mechanism (Zong et al.,
2019). A subsequent yeast two-hybrid screen searching for novel
AXIN1-interacting proteins (Zhang et al., 2014) identified p18/
LAMTOR1, a protein anchored to the lysosomal membrane by
N-terminal myristoyl and palmitoyl modifications (Nada et al.,
2009). p18/LAMTOR1 is a key component of the Ragulator com-
plex, which (as will be discussed later) plays a central role in the
activation of mTORC1 via interaction with the vacuolar ATPase
(v-ATPase) (Bar-Peled et al., 2012; Sancak et al., 2010; Zoncu
et al., 2011). In LAMTOR1 null cells or cells with knockdown of
the v0c subunit of the v-ATPase, AMPK activation induced by
glucose starvation was no longer observed. In addition, AXIN1,
in complex with LKB1, was found to translocate to the lysosomal
surface, forming a supramolecular complex with the Ragulator
and v-ATPase, which was not observed in LAMTOR1 null cells or
cells with knockdown of the v-ATPase v0c subunit (Zhang et al.,
2014). By this mechanism, LKB1 is brought to the vicinity of a
pool of AMPK that appears to permanently reside on the lysosomal
membrane due to N-terminal myristoylation of the b subunit. This
overall mechanism is now referred to as the lysosomal AMPK acti-
vation pathway (Figure 3).
It should be noted that AXIN has two isoforms, AXIN1 and
AXIN2, which are functionally redundant both in Wnt signaling
(Chia and Costantini, 2005) and in the lysosomal AMPK activa-
tion pathway (Zong et al., 2019). While AXIN1 is ubiquitously ex-
pressed, AXIN2 is mainly expressed in neuronal cells and some
actively proliferating cells. For example, AXIN2 is not expressed
in differentiated hepatocytes (Zong et al., 2019), except for a
small population of self-renewing cells adjacent to the central
vein in the liver lobule (Wang et al., 2015a). Similarly, while mouse
embryo fibroblasts (MEFs) only express AXIN1, AXIN2 is also ex-
pressed in HEK293T cells so that if AXIN1 expression is knocked
out in HEK293T cells, the lysosomal AMPK activation pathway
remains intact (Zong et al., 2019). In addition, in some cell types
that rely on glycolysis for ATP production, glucose starvation
may also activate AMPK by the canonical AMP-dependent
pathway, rendering the lysosomal activation pathway redun-
dant. For example, in HEK293 cells (unlike in MEFs), there are
rapid increases in cellular AMP:ATP and ADP:ATP ratios after
glucose removal even when an alternative carbon source like
glutamine is provided (Zhang et al., 2017). In these cells, the ca-
nonical AMP-dependent pathway for AMPK activation operates
independently of the lysosomal AMP-independent pathway in
response to glucose starvation (Zong et al., 2019). Thus, studies
of the lysosomal pathway in some cell types or tissues need to
take into account the possibility not only of expression of
AXIN1 or AXIN2 but also of changing AMP levels.
Although the results of Zhang et al. (2014) demonstrated that
glucose starvation activated AMPK via the lysosomal pathway
in mammals, it remained unclear how the presence or absence
of glucose was sensed. Pursuing this, it became apparent that
aldolase, the glycolytic enzyme that converts FBP (fructose-
1,6-bisphosphate) into triose phosphates, which can also be
associated with the v-ATPase complex, is a direct (physical)
sensor for FBP. When aldolase is unoccupied by FBP (whose
levels rapidly decrease upon glucose deprivation), the v-
ATPase complex undergoes conformational changes that
inhibit its activity as a proton pump (as suggested by increased
pH levels in the lysosomal lumen; Zhang et al., 2017) and also
allow the AXIN1:LKB1 complex to interact with the v-ATPase
and Ragulator. Multiple lines of evidence support the idea
that aldolase is the direct sensor. First, knockdown of all iso-
forms of aldolase caused constitutive activation of AMPK,
even in high glucose. Second, in cells expressing the D34S
mutant of aldolase, which has a greatly reduced kcat despite
an almost unchanged Km for FBP (Morris and Tolan, 1993)
(meaning that FBP will accumulate in the active site of aldolase
even in low glucose), AMPK was not activated by glucose star-
vation (Zhang et al., 2017). Importantly, this mechanism for
AMPK regulation by glucose can occur in the absence of any
changes in adenine nucleotide ratios. For example, in MEFs
transferred from media with high glucose (25 mM) to medium
containing glucose concentrations around 5 mM or in livers of
mice starved overnight (when blood glucose dropped from 9
to 3 mM), AMPK was activated without any associated
changes in cellular AMP:ATP or ADP:ATP ratios. Interestingly,
however, if glutamine (the other major carbon source in the me-
dium) was removed from the medium as well as glucose, there
was an additional, delayed (but ultimately larger) activation of
AMPK that did correlate with increases in AMP:ATP and AD-
P:ATP ratios (Zhang et al., 2017). These results indicate that
the non-canonical glucose-sensing mechanism for AMPK acti-
vation can act in parallel with the canonical AMP-dependent
mechanism. In line with the concept that glucose availability
can be sensed independently of cellular energy status, neither
pyruvate nor glutamine, which both feed into the TCA cycle for
ATP production, prevents lack of glucose from activating
AMPK. Indeed, it is now clear that the AXIN/lysosome-depen-
dent and AMP-dependent mechanisms can co-exist, with their
Cell Metabolism 31, March 3, 2020 477

contributions to overall AMPK activation depending on the
magnitude of any increases in AMP as well as the subcellular
location (Zong et al., 2019).
Another recent study has uncovered the mechanism that
signals the presence or absence of FBP in the active site of
aldolase to the formation of the AXIN-LKB1-AMPK complex on
the lysosomal membrane. It was demonstrated that TRPV (tran-
sient receptor potential V) channels located on the ER (endo-
plasmic reticulum) membrane are required for AMPK activation
in response to low glucose. The current model is that aldolase
that is unoccupied by FBP interacts with TRPV at lysosome:ER
contact sites, inhibiting its Ca2+-releasing activity. Once the
Ca2+ concentrations at the ER-lysosome contact sites falls
below a certain level, TRPV gains affinity for the v-ATPase, re-
configuring its association with aldolase and causing the forma-
tion of the AXIN-based complex to activate AMPK (Li et al.,
2019a). It should be pointed out that the concentration of the
TRPV-released Ca2+ (<1 mM) is well below that required for acti-
vation of CaMKK2, which is not involved in the lysosomal AMPK
activation mechanism. It has been proposed that the pool of
Ca2+ at the ER-lysosome contact sites acts as a kind of buffer
or damper, smoothing the output and thus preventing fluctua-
tions in AMPK caused by rapid oscillations of FBP binding in
the active site of aldolase (Li et al., 2019a).
Glucose starvation also causes rapid activation of the Snf1
complex in S. cerevisiae (Wilson et al., 1996; Woods et al.,
1994), and, intriguingly, complexes containing Sip1 (one of three
b subunit orthologs in yeast) translocate to the vacuolar mem-
brane upon glucose removal (Vincent et al., 2001). However,
the detailed mechanism appears to be different from that in
mammalian cells, because no clear AXIN orthologs are found
in yeast. Once activated, the Snf1 complex phosphorylates the
transcriptional repressor Mig1 (Smith et al., 1999; Treitel et al.,
1998), triggering both its inactivation (Papamichos-Chronakis
et al., 2004) and nuclear export (DeVit and Johnston, 1999).
Mig1 binds to and inhibits the promoters of many glucose-
repressed genes, including the SUC2 gene encoding a secreted
invertase that is required to metabolize alternate carbon sour-
ces, such as sucrose or raffinose (Hedbacker and Carlson,
2008). As in mammalian cells, the Snf1 complex also phosphor-
ylates and inactivates acetyl-CoA carboxylase, potentially inhib-
iting fatty-acid biosynthesis under glucose-limiting conditions
(Mitchelhill et al., 1994; Woods et al., 1994).
Although the effects of starvation for a carbon source are less
well studied in plants, knockout or silencing of the genes encod-
ing the AMPK-a orthologs in the moss Physcomitrella patens
(Thelander et al., 2004) and the higher plant Arabidopsis thaliana
(Baena-González et al., 2007) causes failure to respond appro-
priately to periods of darkness, the equivalent of starvation in
plants. In cells of A. thaliana, the AMPK-a ortholog KIN10 is
responsible for triggering extensive reprogramming of transcrip-
tion affecting thousands of genes, some of which are required for
adaptive responses like starch breakdown during starvation
(Baena-González et al., 2007; Baena-González and Sheen,
2008). SnRK1 complexes also phosphorylate and inactivate
both sucrose phosphate synthase and HMG (3-hydroxy-3-meth-
ylglutaryl)-CoA reductase, potentially inhibiting the anabolic
pathways of sucrose and sterol synthesis (Nukarinen et al.,
2016; Sugden et al., 1999b).
478 Cell Metabolism 31, March 3, 2020
Cell Metabolism
Review
Since activation by starvation for key carbon sources (espe-
cially glucose) appears to be a common feature of the AMPK or-
thologs from mammals, plants, and budding yeast, yet they differ
in their regulation by adenine nucleotides, it is tempting to spec-
ulate that sensing of glucose rather than energy may have been
the ancestral role of the kinase. However, it remains unclear
exactly how carbon starvation causes activation of the orthologs
in plants and yeast.
Downstream Targets of AMPK
AMPK phosphorylates downstream targets containing well-
defined recognition motifs, and at least 60 have now been well
validated—a full discussion of these is beyond the scope of
this article, and readers are referred to a previous review (Hardie
et al., 2016). In general, AMPK phosphorylates and activates
proteins involved in catabolic pathways, thus enhancing ATP
synthesis, while phosphorylating and inactivating proteins
involved in anabolic (biosynthetic) pathways, thus inhibiting cell
growth while conserving ATP. AMPK also causes a cell cycle ar-
rest in G1 phase (Fogarty et al., 2016; Imamura et al., 2001),
although in that case, the direct downstream targets responsible
for the effect are not clear. In this section, we will mention only a
few key targets that are important for the effects of AMPK on
catabolic and anabolic pathways.
Starting with effects on catabolism, in many cell types, AMPK
activation increases glucose uptake via effects on the trafficking
of the glucose transporters, GLUT1 (Barnes et al., 2002) or
GLUT4 (Kurth-Kraczek et al., 1999). This is achieved in part via
phosphorylation and consequent degradation of TXNIP, an a-ar-
restin family member that normally promotes reuptake of GLUT1
and GLUT4 from the plasma membrane by endocytosis (O’Don-
nell and Schmidt, 2019; Wu et al., 2013). In the case of GLUT4,
AMPK also phosphorylates TBC1D1, a GTPase activating pro-
tein (GAP) for members of the Rab family, causing dissociation
of TBC1D1 from intracellular GLUT4-storage vesicles (GSVs)
with consequent conversion of Rabs to their GTP-bound forms,
thus promoting trafficking of GSVs to the plasma membrane
(Pehmøller et al., 2009). AMPK can also phosphorylate and acti-
vate 6-phosphofructo-2-kinase, the enzyme that generates fruc-
tose-2,6-bisphosphate, a potent allosteric activator of the key
glycolytic enzyme 6-phosphofructo-1-kinase. However, this ef-
fect is cell-type-dependent because only the PFKFB2 (Marsin
et al., 2000) or PFKFB3 (Marsin et al., 2002) isoforms of 6-phos-
phofructo-2-kinase, which are not expressed ubiquitously, are
direct targets for AMPK. AMPK also acutely promotes fatty-
acid oxidation by phosphorylating and inactivating the mito-
chondrial isoform of ACC2 (acetyl-CoA carboxylase-2),
thus reducing the local pool of malonyl-CoA, an inhibitor of up-
take of fatty acids into mitochondria via the transport system
involving carnitine palmitoyl-CoA transferase-1 (Winder and Har-
die, 1996).
In the longer term, AMPK activation tends to promote the
oxidative metabolism typical of quiescent cells, rather than the
rapid glucose uptake and glycolysis typical of cells undergoing
rapid proliferation, including tumor cells. First, it promotes mito-
chondrial biogenesis (Zong et al., 2002) as well as expression of
oxidative enzymes (Winder et al., 2000), perhaps by direct phos-
phorylation (Ja €ger et al., 2007) or deacetylation (Cantó et al.,
2009) of the transcriptional co-activator, PGC-1a. Second,
AMPK maintains the cellular content of functional, healthy
Cell Metabolism
Review
mitochondria by promoting both mitophagy, via phosphorylation
of the autophagy kinase ULK1 (Unc-51-like kinase 1) (Egan et al.,
2011b), and mitochondrial fission, perhaps via phosphorylation
of proteins involved in mitochondrial fission like MFF (mitochon-
drial fission factor) or MTFR1L (mitochondrial fission regulator-1-
like) (Ducommun et al., 2015; Schaffer et al., 2015; Toyama et al.,
2016). Because mitochondria can exist in cells as elongated
branching networks that can be of lengths close to that of the
cell diameter, mitochondrial fission may be necessary to break
these networks down into smaller segments suitable for mitoph-
agy. Consistent with this, the phenotypes of muscle-specific
double knockouts of a1/a2 (Lantier et al., 2014) or b1/b2 (O’Neill
et al., 2011) in mice include exercise intolerance associated with
the appearance in electron micrographs of mitochondria of
abnormal size and morphology.
Along with these effects on catabolism, AMPK acutely
switches off most anabolic pathways. It was discovered for its
ability to phosphorylate and inactivate ACC1 (acetyl-CoA
carboxylase-1) and HMG-CoA reductase, two key enzymes of
fatty acid and cholesterol synthesis, respectively (Hardie et al.,
1989). Indeed, phosphorylation of ACC1 at Ser80 (Ser79 in ro-
dents), monitored using phosphospecific antibodies, remains
the most widely used biomarker for AMPK activation in intact
cells. Moreover, mice with knock-in Ser/Ala mutations of the
AMPK sites on ACC1 and ACC2 (Fullerton et al., 2013) or
HMG-CoA reductase (Loh et al., 2018) have elevated levels of tri-
glycerides and cholesterol, respectively, demonstrating that
these phosphorylation sites have regulatory significance in vivo.
AMPK also switches off glycogen synthesis via phosphorylation
of the GYS1 (Jørgensen et al., 2004) and GYS2 (Bultot et al.,
2012) isoforms of glycogen synthase, nucleotide synthesis via
phosphorylation of the PRPS-1 and –2 isoforms of phosphoribo-
syl pyrophosphate synthetase (Qian et al., 2018), and ribosomal
RNA synthesis via phosphorylation of TIF-1A/RRN3, a transcrip-
tion factor for RNA polymerase-1 (Hoppe et al., 2009). Finally,
AMPK switches off the elongation step of protein synthesis in
part via phosphorylation of elongation factor-2 kinase (Johanns
et al., 2017), an atypical Ca2+-dependent kinase that phosphor-
ylates elongation factor-2 and causes pausing in elongation.
Other effects on protein synthesis are mediated indirectly by
inactivation of mTORC1, which is discussed in more detail in a
separate section below.
Yang – the Structure and Regulation of TOR Complexes
Subunit Structure and Evolution
TOR is a serine/threonine protein kinase belonging to the PIKK
(phosphatidylinositol kinase-related kinase) family, which also
includes DNA-PK and ATM (Keith and Schreiber, 1995). TOR is
conserved in all eukaryotes except (as for AMPK) in the case
of a few obligate intracellular parasites like E. cuniculi and
P. falciparum (Tatebe and Shiozaki, 2017; van Dam et al.,
2011) (Figure 1), which may be able to exploit TOR signaling in
the host cell. Whereas most eukaryotes contain a single TOR
gene, a few possess more than one; for example, budding yeast
(S. cerevisiae) and fission yeast (S. pombe) have two (Shertz
et al., 2010) (Table 1), while trypanosomes have up to four
(Saldivia et al., 2013). Early eukaryotes presumably possessed
a single TOR gene that was duplicated and/or lost multiple times
during evolution (Shertz et al., 2010).
TOR was originally identified genetically in S. cerevisiae via
mutations that render cells resistant to the growth-inhibitory
properties of the antibiotic rapamycin (Heitman et al., 1991;
Kunz et al., 1993). It was identified in mammalian cells shortly
thereafter (Brown et al., 1994; Chiu et al., 1994; Sabatini et al.,
1994; Sabers et al., 1995), and the name mTOR (mammalian
TOR) was eventually adopted based on the yeast precedent.
More recently, the HUGO Gene Nomenclature Committee
changed the definition of the mTOR acronym to ‘‘mechanistic
TOR’’ in order to create a common nomenclature for TOR in ver-
tebrates (Hall, 2013). However, this has led to TOR from nema-
todes or even yeast sometimes being referred to as mTOR.
TOR forms two structurally and functionally distinct multipro-
tein complexes termed TOR complexes 1 and 2 (TORC1 and
TORC2), of which only TORC1 is acutely sensitive to rapamycin
(Loewith et al., 2002). The two TOR complexes, like TOR itself,
are conserved from yeast to humans, although TORC1 appears
to be absent from ciliates and TORC2 from plants (Tatebe and
Shiozaki, 2017; van Dam et al., 2011) (Figure 1). In mammals,
mTOR and the adaptor protein mLST8 (mammalian lethal with
SEC13 protein 8) are common to both TOR complexes. RAPTOR
(regulatory-associated protein of TOR) is the defining subunit of
mTORC1, whereas RICTOR (rapamycin-insensitive companion
of mTOR) and mSIN1 (stress-activated MAP kinase interacting
protein 1) define mTORC2.
The domain organization of TOR is also conserved. The C-ter-
minal half of TOR contains a FAT (FRAP, ATM, and TRRAP)
domain followed by the FRB (FKBP-rapamycin binding) domain,
the catalytic kinase domain, and a C-terminal FAT domain
termed FATC. Structural biologists often refer to the FAT, FRB,
kinase, and FATC domains collectively as the FATKIN region
(Baretic et al., 2016; Imseng et al., 2018) (Figure 4). FATKIN re-
gions are found in all PIKK family members, although only the
FRB domain in TOR binds the FKBP-rapamycin complex. All
PIKKs contain long, N-terminal extensions that serve as docking
surfaces for binding partners. The N-terminal half of TOR con-
sists of tandem arrays of HEAT (Huntingtin, elongation factor 3,
PP2A, and TOR) and TPR (tetratricopeptide) repeats. The
HEAT repeats of mTOR bind RAPTOR (Hara et al., 2002; Kim
et al., 2002), which also has several characteristic regions: the
RAPTOR N-terminal conserved (RNC) CASPase-like domain, a
central set of seven ⍺-helical repeats termed the armadillo
(ARM) domain, and a C-terminal seven-bladed WD40 b-propel-
ler (Hara et al., 2002; Kim et al., 2002). By contrast, mLST8 is a
small protein consisting entirely of a WD40 b-propeller.
Structure of the mTORC1 Complex
TORC1 architecture was solved by a combination of X-ray
crystallography and cryo-EM (cryo-electron microscopy) on
truncated mTOR-mLST8 (Yang et al., 2013), RAPTOR from the
fungus Chaetomium thermophilum (Aylett et al., 2016) or the
plant A. thaliana (Yang et al., 2017), and TOR-Lst8 from the fun-
gus Kluyveromyces marxianus (Baretic  et al., 2016). These
studies described mTORC1 at 4.4  A (Yang et al., 2016) and 3.0

A resolution (Yang et al., 2017), mTORC1 in complex with
FKBP-rapamycin at 5.9  A (Aylett et al., 2016), and mTORC1
bound to its activator RHEB at 3.4  A (Yang et al., 2017).
mTORC1 is a 1 MDa homodimer of heterotrimers (each of the
latter containing mTOR, RAPTOR, and mLST8) that adopts a
rhomboidal (lozenge) shape with a large central cavity (Figure 4).
Cell Metabolism 31, March 3, 2020 479
 Cell Metabolism

Review

Figure 4. Human mTORC1 Architecture

(A) Linear representation of the domain organization of mTOR, RAPTOR, and mLST8. The residue numbers indicate the domain boundaries. Grey areas in
RAPTOR indicate regions presumed to be disordered linkers, comprising amino acids 687–805 and 841–949.
(B) Cryo-EM derived model of human mTORC1 (PDB: 6BCX) (Yang et al., 2017), with domains colored according to the primary structure scheme in (A). Key
residues for mTORC1 activation at the catalytic site (Asp2338, His2340, Asn2343, and Asp2357; Yang et al., 2013) are highlighted in red, while the two copies of
the TOS peptide of 4EBP are shown in purple. A gray line indicates the RAG binding region. Gray dashed lines represent the two disordered linker regions in
RAPTOR. AMPK, PKA, and NLK phosphorylate RAPTOR at Ser722 plus Ser792, Ser791, and Ser863, respectively. EP300 acetylates RAPTOR at Lys1097
(residue highlighted in magenta). RHEB binds the N terminus and FAT domain of mTOR, distal to the catalytic site (not shown). See main text for details.

It exhibits two-fold (C2) symmetry with the axis of symmetry
passing through the central cavity. The FATKIN region of each
of the two copies of mTOR forms a compact unit located near
the central cavity, on opposite sides of the C2 axis. The two FAT-
KIN regions come close to each other but make little or no con-
tact. Each kinase site is located at the bottom of a deep catalytic
cleft that is partly obscured by surrounding structural elements,
suggesting that the kinase activity is regulated by physically re-
stricting access to the catalytic site (Yang et al., 2017; Yang
et al., 2013). The HEAT repeats of each mTOR subunit form
two distinct helical solenoids, one a low-curvature bridge/M-
HEAT (hereafter referred to as the ‘‘bridge’’) and the second a
high-curvature horn/spiral/N-HEAT (hereafter referred to as the
‘‘horn’’) peripherally linked to the bridge (Aylett et al., 2016; Bare-
 et al., 2016; Yang et al., 2017). The horn of one copy of mTOR
tic
packs against the bridge of the other to mediate dimerization and
form the central cavity. The two-fold symmetry is likely
conserved among TORC orthologs because (1) there is a high
degree of conservation throughout the HEAT repeat region of
TOR and (2) TOR from K. marxianus (Baretic  et al., 2016) and hu-
mans (Aylett et al., 2016) are architecturally identical. The horn
and bridge, in addition to forming the dimer interface, are
exposed, suggesting an additional role in binding regulatory or
accessory proteins. mLST8 binds to the kinase domain of
mTOR and thereby constitutes the ends of the short axis of the
mTORC1 rhomboid. RAPTOR has an extended Z-like shape
with the RNC domain and WD40 b-propeller located at opposite
ends, connected by the ARM domain (Aylett et al., 2016; Yang
et al., 2017). RAPTOR also contributes to the mTORC1 dimer
interface, because the ARM domain of one RAPTOR binds the
horn of one mTOR molecule and the bridge of the other, thereby
linking the two copies of mTOR. The RAPTOR b-propeller do-
mains are at the ends of the long axis of mTORC1.
Importantly, RAPTOR is also required for mTORC1 substrate
recruitment. The region in RAPTOR responsible for substrate
binding is in a cleft between the RNC and the ARM domains,
located z65  A from the catalytic site (Figure 4) (Yang et al.,
2017), via which RAPTOR binds a sequence of five amino acids

480 Cell Metabolism 31, March 3, 2020

Cell Metabolism
Review
termed the TOS (TOR signaling) motif. The TOS motif is defined
as FXV[E/D]V, where V is a hydrophobic residue and X any res-
idue (Gouw et al., 2018; Nojima et al., 2003; Schalm and Blenis,
2002; Yang et al., 2017). TOS motifs are present in some TORC1
substrates, such as ribosomal protein S6 kinase (S6K; TOS motif
FDIDL) and eukaryotic translation initiation factor 4E binding pro-
tein (4EBP; TOS motif FEMDI) (Nojima et al., 2003; Schalm and
Blenis, 2002; Schalm et al., 2003). However, the mTORC1 sub-
strates ULK1 (Dunlop and Tee, 2013) and TFEB (transcription
factor EB) (Roczniak-Ferguson et al., 2012; Settembre et al.,
2012) interact with RAPTOR yet lack an obvious TOS motif.
Furthermore, although the TOS-binding region of RAPTOR is
highly conserved from yeast to mammals, TORC1 substrates
in lower eukaryotes seem to lack TOS motifs, so that it is unclear
how TORC1 recognizes its substrates in those organisms.
Inhibition of TOR by rapamycin depends on the formation of a
complex between rapamycin and the cytoplasmic immunophilin
FKBP12 (FK506-binding protein of 12 KDa) (Benjamin et al.,
2011). An FKBP-rapamycin complex binds the FRB domain at
the lip of the TOR catalytic cleft, forming a lid that physically pre-
vents access of substrates to the catalytic site. FKBP-rapamycin
does not induce a conformational change in mTOR, suggesting
that FKBP-rapamycin indeed acts by obstructing substrate ac-
cess (Aylett et al., 2016; Yang et al., 2017; Yang et al., 2013).
TORC2 is not acutely inhibited by rapamycin, because the
FKBP-rapamycin binding site in the TOR FRB domain in
TORC2 is masked by RICTOR (Chen et al., 2018; Gaubitz
et al., 2015; Karuppasamy et al., 2017; Stuttfeld et al., 2018).
Cryo-EM studies have resolved S. cerevisiae (Karuppasamy
et al., 2017) and human (Chen et al., 2018; Stuttfeld et al.,
2018) TORC2 at intermediate resolution. The two mTOR com-
plexes share many features, including C2 symmetry, similar
binding sites for RAPTOR and RICTOR, and a deep catalytic
cleft. However, full structural interpretation of mTORC2 awaits
higher resolution structural data.
Regulation of mTORC1 by Lysosomal Recruitment and
Growth Factors
TOR controls cell growth and metabolism in response to nutri-
ents, growth factors, and (in part through AMPK) cellular energy
status. Nutrients, especially amino acids, are likely to be the
ancestral TORC1 activating inputs, as they are sufficient to acti-
vate TORC1 in unicellular organisms such as yeast. However, in
multicellular organisms, TORC1 activation requires additional
input from growth factors. Mechanistically, amino acid and
growth factor inputs converge on mTORC1 as follows: (1) amino
acids stimulate translocation of mTORC1 from the cytosol to the
lysosome, where they encounter the small G protein RHEB (RAS
homolog enriched in brain), and (2) growth factors activate lyso-
somal RHEB, enabling it to activate mTORC1 in turn (see below).
Amino acid availability is transmitted to TORC1 mainly via the
RAGs (Ras-related family of small GTPases) (González and Hall,
2017; Nicastro et al., 2017; Wolfson and Sabatini, 2017)
(Figure 5). There are four RAGs in mammals (RAGA through
RAGD) and two in S. cerevisiae (Gtr1 and Gtr2) that form obligate
heterodimers of RAGA or RAGB with RAGC or RAGD and Gtr1
with Gtr2. RAGs are attached to the lysosome in mammalian
cells through the pentameric Ragulator complex (Bar-Peled
et al., 2012; Sancak et al., 2010), while the Gtr1-Gtr2 heterodimer
is attached to the vacuole in yeast through the trimeric Ego com-
plex (Kogan et al., 2010; Levine et al., 2013; Powis et al., 2015;
Zhang et al., 2012). Clearly, the lysosome or vacuole is the
TORC1 signaling hub in all eukaryotic cells. Amino-acid suffi-
ciency promotes the TORC1-activating conformation of the
RAG-Gtr heterodimer (RAGA/B or Gtr1 loaded with GTP, and
RAGC/D or Gtr2 loaded with GDP). In mammals, the active
RAG heterodimer binds RAPTOR and thereby recruits mTORC1
from the cytosol to the lysosomal surface, while in budding
yeast, TORC1 is constitutively bound to the vacuolar surface,
and the active Gtr1-Gtr2 heterodimer binds Kog1 (yeast ortholog
of RAPTOR) to stimulate TORC1 via an unknown mechanism
(Binda et al., 2009). From yeast two-hybrid experiments, it has
been proposed that a region of Kog1 comprising amino acids
777–814 in the central ARM domain interacts with Gtr1 (Sekigu-
chi et al., 2014). The region in Kog1 is conserved in RAPTOR
(amino acids 777–814 in Kog1 correspond to amino acids 595–
632 in RAPTOR). Consistent with this, recent structural analyses
of RAGAGTP-RAGCGDP in complex with mTORC1 (Anandapada-
manaban et al., 2019) or with RAPTOR-Ragulator (Rogala et al.,
2019) revealed that the region in RAPTOR comprising amino
acids 546–650 binds RAGAGTP. Two additional regions of
RAPTOR, located between the ARM and WD40 b-propeller do-
mains, interact with RAGCGDP (Rogala et al., 2019). One region
comprises amino acids 795–806 and the other, amino acids
916–937. The last has been referred to as the ‘‘RAPTOR claw’’
due to its shape (Rogala et al., 2019). Interestingly, it has been
suggested that the stress-activated MAP kinase-related kinase
NLK (Nemo-like kinase) phosphorylates RAPTOR at Ser863,
thereby disrupting RAG-RAPTOR interaction and inhibiting
mTORC1 (Yuan et al., 2015). Ser863 is in a structurally unsolved
and thus presumably disordered linker region (residues 841 to
949) between the ARM and WD40 b-propeller domains that con-
tains several phosphorylation sites (Foster et al., 2010; Wang
et al., 2009) (Figure 4) (see below).
The nucleotide binding status of the RAGs is tightly regulated
by conserved GAPs (GTPase activator proteins) and GEFs (gua-
nine nucleotide exchange factors) (González and Hall, 2017; Ni-
castro et al., 2017; Wolfson and Sabatini, 2017) (Figure 5). The
heterotrimeric GATOR1 (GAP activity toward RAGs-1) complex
is the GAP for RAGA/B and thus negatively regulates mTORC1
activity (Bar-Peled et al., 2013; Panchaud et al., 2013a; Shen
et al., 2018; Shen et al., 2019). GATOR1 is tethered to the lyso-
somal surface by KICSTOR (KPTN, ITFG2, C12orf66, and
SZT2-containing regulator of mTORC1) (Peng et al., 2017; Wolf-
son et al., 2017). The heteropentameric GATOR2 complex acti-
vates mTORC1 by binding and negatively regulating GATOR1
via an undefined mechanism (Bar-Peled et al., 2013; Panchaud
et al., 2013b). The lysosomal amino-acid transporter SLC38A9
(Jung et al., 2015; Rebsamen et al., 2015; Wang et al., 2015b;
Wyant et al., 2017) acts as a GEF for RAGA (Shen and Sabatini,
2018). The Ragulator complex, which was initially described as
the GEF for RAGA/B (Bar-Peled et al., 2012), is now proposed
instead to activate mTORC1 by accelerating the release of
GTP from RAGC (Shen and Sabatini, 2018), while the identity
of the GEF for RAGC/D remains unclear. FLCN (folliculin),
together with its binding partners FNIP1 and FNIP2 (folliculin-in-
teracting protein 1 and 2), has been identified as the GAP for
RAGC/D and thus positively regulates mTORC1 (Petit et al.,
2013; Tsun et al., 2013).
Cell Metabolism 31, March 3, 2020 481
 Cell Metabolism

Review

Figure 5. Cross-Talk between mTORC1 and AMPK Signaling Pathways in Mammals

Proteins shown in green promote activation of mTORC1 (blue box), while proteins shown in red promote its inhibition. Inputs into mTORC1 from AMPK signaling
are shown in gray, because AMPK and mTORC1 would not be simultaneously active. Dashed lines indicate indirect interactions. Amino acids and growth factors
activate mTORC1, which then promotes cell growth by stimulating anabolic processes. Growth-factor-stimulated PI3K activates mTORC2 (yellow box) by
promoting its association with the ribosome. Active mTORC2 then promotes cell proliferation and survival. See main text for details.

Upon amino-acid starvation, the RAG heterodimer assumes
an inactive configuration (RAGA/B loaded with GDP and
RAGC/D with GTP) that is unable to recruit mTORC1 to the lyso-
somal surface so that mTORC1 remains cytosolic and inactive. It
has been proposed also that the ‘‘inactive’’ conformation of the
RAG heterodimer recruits TSC2 (tuberous sclerosis complex 2)
to the lysosome to inhibit mTORC1 (Demetriades et al., 2014;
Demetriades et al., 2016; Menon et al., 2014). In budding yeast,
glucose withdrawal triggers a Gtr-dependent formation of a vac-
uole-associated cylindrical filament of TORC1 molecules,
termed a TOROID (TORC1 organized in inhibited domains).
TOROID formation leads to TORC1 inactivation, and low-resolu-
tion cryo-EM reconstructions suggest that this oligomerization
causes steric occlusion of the TORC1 active site (Prouteau
et al., 2017). It is not known whether mTORC1 forms TOROID-
like structures.
As discussed in the introduction to this review, it is thought
that growth factor signaling co-evolved with multicellularity, at
which time it was grafted onto the ancestral nutrient-activated
TORC1 signaling pathway (Ben-Sahra and Manning, 2017; Guri
and Hall, 2016; Kim and Guan, 2019). Growth factors like insulin
bind to RTKs (receptor tyrosine kinases) to activate PI3K (phos-
phatidylinositol-4,5-bisphosphate 3-kinase), thereby gener-
ating PIP3 (phosphoinositide 3, 4, 5-trisphosphate) (Figure 5).
PIP3 then co-recruits PDK1 (phosphoinositide-dependent ki-
nase-1) and AKT via their PIP3-binding PH (pleckstrin homol-
ogy) domains to the plasma membrane, where PDK1 phos-
phorylates Thr308 in the activation loop of AKT. Activated
AKT in turn phosphorylates TSC2 on multiple sites to induce
the release of the heterotrimeric TSC complex from the lyso-
some (Inoki et al., 2002; Menon et al., 2014). The TSC complex
consists of TSC1, TSC2, and TBC1D7, and it acts as a GAP to-
ward RHEB (Dibble et al., 2012). Reduced TSC complex GAP
activity at the lysosome leads to an increase in activated,
GTP-loaded RHEB, which then binds the N terminus and FAT
domain of mTOR to allosterically realign residues in the cata-
lytic site and activate mTORC1 (Chao and Avruch, 2019; Long
et al., 2005; Yang et al., 2017).

482 Cell Metabolism 31, March 3, 2020

Cell Metabolism
Review
Amino-Acid Sensors
Leucine, arginine, and glutamine are among the most effective
amino acids for activation of mTORC1 (Figure 5). The identity
of the amino acid sensors upstream of TORC1 has begun to
emerge recently (Wolfson and Sabatini, 2017). The cytoplasmic
proteins SESTRIN2 (Chantranupong et al., 2014; Kim et al.,
2015; Parmigiani et al., 2014; Saxton et al., 2016b; Saxton
et al., 2016c; Wolfson et al., 2016) and CASTOR (cellular arginine
sensor for mTORC1) (Chantranupong et al., 2016; Saxton et al.,
2016a; Xia et al., 2016) bind and transmit the availability of
leucine and arginine, respectively, to mTORC1 via the GATOR
complexes. Under conditions of leucine and arginine depriva-
tion, SESTRIN2 and CASTOR1 bind and most likely inhibit
GATOR2 upstream of mTORC1. However, growth-promoting
levels of leucine and arginine disrupt the interactions of SES-
TRIN2 and GATOR2 (Wolfson et al., 2016) and CASTOR1 and
GATOR2 (Saxton et al., 2016a); this releases free GATOR2 and
thereby activates mTORC1 (Figure 5). SESTRINs may also inhibit
mTORC1 by activating AMPK (Lee et al., 2016). However,
budding yeast lacks SESTRIN and CASTOR orthologs (Wolfson
and Sabatini, 2017). Whether, and, if so, how arginine or leucine
availability is transmitted to TORC1 in organisms lacking these
proteins is not known. Leucine and glutamine can also activate
mTORC1 via glutaminolysis and consequent production of a-ke-
toglutarate upstream of RAGs (Durán et al., 2013; Durán et al.,
2012), while glutamine also activates mTORC1 independently
of the RAGs via the small GTPase ARF1 and the v-ATPase (Jew-
ell et al., 2015).
It has been reported that LeuRS (leucyl-tRNA synthetase) acts
as a cytoplasmic leucine sensor to activate mTORC1 via a RAG-
independent mechanism. Leucine-bound LeuRS binds and acti-
vates the class-III phosphoinositide kinase VPS34 that is present
in non-autophagic structures. Active VPS34 stimulates PLD1
(phospholipase D1), thereby increasing phosphatidic acid levels,
which promote lysosomal activation of mTORC1 (Yoon et al.,
2016; Yoon et al., 2011).
In some cell types, such as epithelial, glial, and mesenchymal
stem cells, leucine can activate mTORC1 via production of
acetyl-CoA. Acetyl-CoA stimulates the acetyl transferase
EP300 to acetylate RAPTOR at Lys1097, thereby promoting
mTORC1 activity (Son et al., 2019). The acetylated residue is
located in the WD40 b-propeller of RAPTOR, close to the ARM
domain (Figure 4). It is unclear whether RAPTOR acetylation af-
fects mTORC1 structure.
Finally, methionine signals to mTORC1 through synthesis of
the methyl donor SAM. SAM availability is transmitted to
mTORC1 via SAMTOR (SAM sensor upstream of mTORC1),
with SAM inhibiting the interaction between SAMTOR and
GATOR1, thereby activating mTORC1 (Gu et al., 2017a).
Downstream Targets of mTORC1
TOR promotes cell growth by stimulating anabolic processes,
such as ribosome biogenesis and protein, lipid, and nucleotide
synthesis, while repressing catabolic processes like autophagy
(Ben-Sahra and Manning, 2017; Saxton and Sabatini, 2017; Shi-
mobayashi and Hall, 2014). mTORC1 promotes protein synthesis
by phosphorylating (1) S6K at Thr389 in its hydrophobic motif, to
increase translation initiation and elongation, and (2) 4EBP, to pro-
mote cap-dependent translation. mTORC1 also induces purine
synthesis via the tetrahydrofolate cycle (Ben-Sahra et al., 2016)
and pyrimidine synthesis by phosphorylating and activating
CAD (carbamoyl-phosphate synthetase 2, aspartate transcarba-
mylase, and dihydroorotase) via S6K (Ben-Sahra et al., 2013; Ro-
bitaille et al., 2013). Furthermore, mTORC1 promotes lipogenic
gene expression by activating the SREBP (sterol-regulatory
element-binding protein) transcription factor (Ben-Sahra and
Manning, 2017). mTORC1 also inhibits autophagy by phosphory-
lating the autophagy-inducing kinase ULK1 (Kim et al., 2011) and
TFEB (transcription factor EB) (Martina et al., 2012; Roczniak-Fer-
guson et al., 2012; Settembre et al., 2012). Phosphorylated TFEB
remains cytosolic and inactive, thus failing to induce expression of
genes required for autophagy and lysosome biogenesis (Puertol-
lano et al., 2018) (Figure 5).
S6K has several substrates, including ribosomal protein S6
and insulin receptor substrate 1 (IRS1). Phosphorylation of
IRS1 by S6K inhibits IRS1, thereby forming a negative feedback
loop acting on PI3K and mTORC2 (Shimobayashi and Hall,
2014). mTORC2 regulates cytoskeletal remodeling, proliferation,
and survival by phosphorylating and activating AGC kinase fam-
ily members like AKT at Ser473, PKC (protein kinase C) and SGK
(serum/glucocorticoid-regulated kinase) (Guri and Hall, 2016).
Yin-Yang: Regulation of mTORC1 by AMPK
If the energy status of cells is compromised, it would not be a
sensible idea for them to grow or divide, even if nutrients were
still available. It therefore makes sense that AMPK should switch
off mTORC1. Indeed, AMPK activation switches off the mTORC1
complex by twin mechanisms:
1. AMPK phosphorylates TSC2 at Thr1271 and Ser1387 (res-
idue numbering from human isoform 1 (NP_000539); these
sites are referred to as Thr1227 and Ser1345 in the original
paper (Inoki et al., 2003). Mutation of these two sites was
found to reduce the ability of the glycolytic inhibitor 2-de-
oxyglucose to inhibit S6K and 4EBP phosphorylation. This
phosphorylation is sometimes assumed to promote the
GAP activity of the TSC complex toward RHEB, although
this has not been directly demonstrated.
2. AMPK directly phosphorylates the RAPTOR component of
mTORC1 at two sites, Ser722 and Ser792. Once again,
mutation of these two sites was found to reduce the ability
of the AMPK activators, AICAR or phenformin, to inhibit
S6K and 4EBP phosphorylation (Gwinn et al., 2008),
although the detailed mechanism for this inhibitory effect
remains unclear. Ser722 and Ser792 lie in a structurally un-
characterized, and likely disordered, region within the
RAPTOR ARM domain (residues 687–805) (Figure 4)—
note that some publications incorrectly place Ser792 in
the RAPTOR b-propeller. Curiously, PKA (cyclic AMP-
dependent protein kinase) phosphorylates RAPTOR on
Ser791, but not Ser792, and is reported to either inhibit
(Jewell et al., 2019) or activate (Liu et al., 2016)
mTORC1—the reasons for this discrepancy are not clear.
These mechanisms may be at least partly conserved across
eukaryotes. Although there appear to be no direct orthologs of
TSC2 in either budding yeast or plants, there is evidence that
phosphorylation of the RAPTOR orthologs in S. cerevisiae
(Hughes Hallett et al., 2015) and plants (Nukarinen et al., 2016)
Cell Metabolism 31, March 3, 2020 483

also leads to inactivation of TORC1 in those organisms. While
these effects were dependent upon the AMPK orthologs, neither
of the two well-defined sites for AMPK in mammalian RAPTOR
(Gwinn et al., 2008) is conserved in S. cerevisiae, and only one
is conserved in plants. The detailed mechanisms for these ef-
fects may therefore be different.
These results therefore show that activation of mammalian
AMPK inhibits mTORC1 via two mechanisms, equivalent to the
fail-safe method of using both ‘‘belt and braces (suspenders)’’
to hold up one’s trousers (pants)! A major effect of mTORC1 acti-
vation is to promote translation, particularly of mRNAs encoding
proteins required for rapid cell growth, including ribosomal pro-
teins. Since protein synthesis accounts for as much as 20% of
total energy turnover in rapidly growing cells (Buttgereit and
Brand, 1995), switching it off would have a major effect to
conserve energy.
Although it is therefore clear that AMPK inhibits mTORC1, very
recently, it has been reported, rather counter-intuitively, that it ac-
tivates mTORC2 (Kazyken et al., 2019). Treatment of serum-
deprived mouse embryo fibroblasts, HEK293 cells, or primary
mouse hepatocytes with AMPK activators like AICAR, biguanides,
or A-769662 was found to increase phosphorylation of the
mTORC2 site on AKT, Ser473. Although these activators all
have known ‘‘off-target’’ (i.e., AMPK-independent) effects, and
more specific AMPK activators are now available, their effects
were reduced, although not eliminated, in cells with AMPK
knocked out or knocked down, suggesting that they were at least
partly mediated by AMPK. The effects were associated with phos-
phorylation of Ser1261 on mTOR and unidentified site(s) on RIC-
TOR, although Ser1261 phosphorylation did not appear to be
required for enhanced phosphorylation of AKT. The authors pro-
posed that the activation of mTORC2 by AMPK represents part
of the mechanism by which the latter increases cell survival during
energetic stress and in some circumstances may therefore para-
doxically promote tumorigenesis (Kazyken et al., 2019).
In addition, there seems to be a dual ‘‘belt and braces’’ system
to turn off mTORC1 when cells are facing shortage of glucose
supply. Besides the above-mentioned mechanisms involving
phosphorylation of mTORC1-related factors by AMPK, glucose
deprivation can inactivate mTORC1 independently of AMPK. Mu-
tations of RAGA/B that abolish GTPase activity completely abro-
gated inhibition of mTORC1 by glucose starvation, despite intact
activation of AMPK, suggesting that RAGs or RAG-interacting
partners may play a more direct role in controlling mTORC1 in
response to nutrients (Efeyan et al., 2013; Kalender et al., 2010).
Indeed, in low glucose, AXIN translocates to the surface of the
lysosome and interacts with the v-ATPase and Ragulator, thereby
facilitating the release of mTORC1 from the lysosomal surface
(Zhang et al., 2014). Additional evidence for AMPK-independent
regulation is that mTORC1 suppression after glucose starvation
occurs several hours later in AXIN null compared to AXIN wild-
type cells in which AMPKa1/a2 had been knocked out (Zhang
et al., 2014). This additional device highlights the importance of in-
hibiting mTORC1 when glucose is absent.
Antagonistic Effects of AMPK and mTORC1 on
Autophagy and Lysosome Biogenesis
Autophagy, of which mitophagy (discussed above) is a special
case, is the process by which cellular contents that are surplus
484 Cell Metabolism 31, March 3, 2020
Cell Metabolism
Review
to requirements are engulfed into lysosomes, where they
are broken down to recycle their components for catabolism or
re-use. By phosphorylating the autophagy-initiating kinase
ULK1 at distinct sites, AMPK activates while mTORC1 inhibits
autophagy (Egan et al., 2011b; Kim et al., 2011). AMPK can
therefore promote autophagy not only by direct phosphorylation
of ULK1 but also indirectly by inactivating mTORC1 via mecha-
nisms discussed in the previous paragraph.
One key downstream target of ULK1 is BECLIN-1, which
forms a complex with VPS34, a class-III phosphoinositide kinase
that generates phosphoinositide-3-phosphate (PI3P) on intracel-
lular membranes. PI3P recruits to those membranes proteins
containing PI3P-binding domains, which mediate subsequent
membrane-trafficking events. VPS34 occurs in several distinct
complexes; AMPK appears to activate complexes involved in
autophagy by phosphorylating BECLIN-1, while inhibiting those
involved in other membrane-trafficking events by phosphory-
lating VPS34 itself; this switch depends on the presence of
ATG14L in the former complex (Kim et al., 2013). Thus, AMPK
may divert membrane traffic (an energy-requiring process) to-
ward the autophagy/mitophagy pathway and away from other
trafficking events that might be a luxury in cells experiencing
glucose starvation or energy stress.
As well as their acute effects on autophagy, in the longer term,
AMPK and mTORC1 also act antagonistically via effects on the
related transcription factors EB and E3 (TFEB and TFE3), which
induce genes involved in lysosome biogenesis and autophagy.
mTORC1 directly phosphorylates TFEB and TFE3, and this pro-
motes their retention in the cytoplasm, inhibiting their transcrip-
tional functions (Puertollano et al., 2018). By contrast, AMPK
activation promotes dephosphorylation and nuclear transloca-
tion of TFEB, an effect that appears to be at least partially inde-
pendent of mTORC1 (Collodet et al., 2019). One possible mech-
anism for increased transcription at TFEB/TFE3-regulated
promoters in response to AMPK activation is the increased
expression of CARM1 (coactivator-associated arginine methyl-
transferase-1) due to downregulation of a E3 ubiquitin ligase
containing SKP2 (S-phase kinase-associated protein-2) (Shin
et al., 2016). Another transcription factor, FOXO3a, is phosphor-
ylated by AMPK at several sites (Greer et al., 2007), and this en-
hances its ability to repress SKP2 expression. The final link in this
proposed chain of events is that CARM1 is recruited to pro-
moters of genes involved in autophagy and lysosome biogenesis
by TFEB, leading to methylation of Arg17 on histone H3 and
consequent activation of transcription at those sites (Shin
et al., 2016).
Yang-Yin: Regulation of AMPK by TORC1 and/or
Upstream Pathways
There is one report that rapamycin treatment of budding yeast, in
wild-type strains but not in strains expressing a TOR1 mutation
that confers rapamycin resistance, increases phosphorylation
of Thr210 in Snf1 (equivalent to Thr172 in mammalian AMPK)
(Orlova et al., 2006). In mouse embryo fibroblasts, neither rapa-
mycin nor the catalytic site inhibitor of mTOR, Torin1, appeared
to affect AMPK activity (Zhang et al., 2014). However, while this
article was in final preparation, Ling et al. (2020) provided evi-
dence, using cells from both S. pombe and mammals, that
TORC1 directly phosphorylates the AMPK-a1 and a2 subunits
Cell Metabolism
Review
at conserved serine residues within the a-linkers, leading to inhi-
bition of AMPK. The detailed mechanism for this effect remains
unclear, because the phosphorylation did not appear to affect
phosphorylation or dephosphorylation of Thr172, or allosteric
activation by AMP, in cell-free assays (Ling et al., 2020). AMPK
can also be downregulated by the upstream insulin signaling
pathway that activates mTORC1. The insulin-stimulated protein
kinase, AKT, phosphorylates Ser496 (human numbering,
Q13131) in the a1 catalytic subunit of AMPK (Horman et al.,
2006), and this downregulates (while not completely abolishing)
AMPK signaling by inhibiting the phosphorylation of Thr172 by
LKB1 (Hawley et al., 2014). Ser496 in AMPK-a1 can also be
phosphorylated by PKC (Heathcote et al., 2016) and PKA (Hurley
et al., 2006). Ser496 occurs in a serine/threonine-rich sequence
just prior to the C-terminal a-helix of AMPK-a1 that has been
termed the ‘‘ST loop’’ (Figure 2). A similar sequence is present
in the a2 isoform, although in that case, the serine residue equiv-
alent to Ser496 (Ser491) is a poor substrate for AKT and appears
to be modified by autophosphorylation instead (Hawley et al.,
2014) (it should therefore not be assumed, as is often done,
that the regulation of the two isoforms by ST loop phosphoryla-
tion is identical). Relevant to this, Ser491 in AMPK-a2 has been
reported to be phosphorylated by S6K1 (Dagon et al., 2012),
which is interesting because the latter is phosphorylated and
activated by mTORC1. However, it is puzzling why there was
no phosphorylation of Ser491 in the absence of S6K1 in this
study (Dagon et al., 2012), when others have observed that
Ser491 in a2 complexes undergoes rapid autophosphorylation
(Hawley et al., 2014).
The ST loop may be subject to multisite phosphorylation,
because GSK3 has been reported to phosphorylate sequentially
within the ST loop of a1 at Thr490, Ser486, and Thr482 (human
numbering, Q13131), which was proposed to promote Thr172
dephosphorylation (Suzuki et al., 2013). Interestingly, the ST
loop is also present in AMPK-a orthologs from C. elegans and
vertebrates but is absent in those from D. melanogaster and
S. cerevisiae, suggesting that it is a regulatory sequence that
has been inserted during evolution. In the currently available
crystal structures of mammalian heterotrimers, the ST loop has
either been deliberately deleted or is not resolved. However,
the residues at either end of the missing loop lie just 20 and
40 Å from Thr172, suggesting that once phosphorylated, the
loop might interact with the kinase domain and physically block
access to Thr172 (Figure 2). Indeed, there is experimental sup-
port for this model (Hawley et al., 2014).
Another potential ‘‘Yang-Yin’’ interaction involves the phos-
phorylation of AMPK by ULK1, the autophagy-regulating kinase
that is inactivated/activated by phosphorylation at distinct sites
by mTORC1/AMPK, respectively (Egan et al., 2011a). ULK1
has been reported to phosphorylate Ser108 on AMPK-b1 but
not -b2 (Dite et al., 2017). Phosphorylation of Ser-108 is known
to stabilize the ADaM site (see above) by interacting with
conserved threonine and lysine residues on the N-lobe of the a
subunit kinase domain (Calabrese et al., 2014; Xiao et al.,
2013) and is required for allosteric activation of AMPK by
ADaM site ligands both with purified AMPK (Scott et al., 2014)
and in intact cells (Dite et al., 2017). However, understanding
the significance of this requires further study, partly because
Ser108 is also rapidly modified by AMPK itself by cis-autophos-
phorylation (Scott et al., 2014) and partly because the natural li-
gands that bind to the ADaM site, if they exist, have not yet been
identified.
Conclusions and Perspectives
We have argued in this review that the AMPK and TOR pathways
arose very early during eukaryotic evolution and may have been
required to regulate cell growth in response to the availability of
the energy or nutrients provided by some of the newly acquired
subcellular compartments, such as mitochondria or lysosomes/
vacuoles. The recent findings that lysosomes/vacuoles repre-
sent key hubs for nutrient sensing by both AMPK and TOR
may reflect the fact that early unicellular eukaryotes utilized
phagocytosis or pinocytosis for feeding, with nutrients being
delivered initially to lysosomes or the vacuole, which, in a unicel-
lular eukaryote, can therefore be considered to be equivalent to
the gut. Just as the gut (and associated endocrine pancreas) of
multicellular animals has become a hub for nutrient sensing
and signaling, so perhaps did the lysosome or vacuole of unicel-
lular eukaryotes.
AMPK can be regarded as representing the Yin or ‘‘dark’’ side
of growth control that is activated by lack of energy or nutrients
and switches off cell growth, while TOR represents the Yang or
‘‘bright’’ side that is activated by availability of nutrients and pro-
motes cell growth. In general, TOR pathways promote anabolic
activities, while AMPK pathways exercise a brake on them.
These pathways clearly act in opposition to each other, and it
is not surprising, as discussed in this review, that there are com-
plex interactions between them. As in Taoist philosophy, the
exquisite balance between Yin and Yang ultimately ensures ho-
meostasis and a healthy cell or organism.
ACKNOWLEDGMENTS
M.N.H. acknowledges support from the European Research Council (MERiC,
609883), the Louis Jeantet Foundation, and the Swiss National Science Foun-
dation (310030B-179569). S.C.L. acknowledges supports from the National
Natural Science Foundation of China (#31730058, #31430094, #31601152
and #31690101) and the National Key R&D Program of China
(2016YFA0502001). D.G.H. acknowledges support from an Investigator Award
from the Wellcome Trust (204766). We thank Stefan Imseng (Biozentrum, Uni-
versity of Basel) for assistance in preparation of Figure 4.
REFERENCES
Alderson, A., Sabelli, P.A., Dickinson, J.R., Cole, D., Richardson, M., Kreis, M.,
Shewry, P.R., and Halford, N.G. (1991). Complementation of snf1, a mutation
affecting global regulation of carbon metabolism in yeast, by a plant protein ki-
nase cDNA. Proc. Natl. Acad. Sci. USA 88, 8602–8605.
Alessi, D.R., Sakamoto, K., and Bayascas, J.R. (2006). LKB1-dependent
signaling pathways. Annu. Rev. Biochem. 75, 137–163.
Anandapadamanaban, M., Masson, G.R., Perisic, O., Berndt, A., Kaufman, J.,
Johnson, C.M., Santhanam, B., Rogala, K.B., Sabatini, D.M., and Williams,
R.L. (2019). Architecture of human Rag GTPase heterodimers and their com-
plex with mTORC1. Science 366, 203–210.
Anashkin, V.A., Baykov, A.A., and Lahti, R. (2017). Enzymes regulated via cys-
tathionine beta-synthase domains. Biochemistry (Mosc.) 82, 1079–1087.
Apfeld, J., O’Connor, G., McDonagh, T., DiStefano, P.S., and Curtis, R. (2004).
The AMP-activated protein kinase AAK-2 links energy levels and insulin-like
signals to lifespan in C. elegans. Genes Dev. 18, 3004–3009.
Aylett, C.H.S., Sauer, E., Imseng, S., Boehringer, D., Hall, M.N., Ban, N., and
Maier, T. (2016). Architecture of human mTOR complex 1. Science 351, 48–52.
Cell Metabolism 31, March 3, 2020 485

Baena-González, E., and Sheen, J. (2008). Convergent energy and stress
signaling. Trends Plant Sci. 13, 474–482.
Baena-González, E., Rolland, F., Thevelein, J.M., and Sheen, J. (2007). A cen-
tral integrator of transcription networks in plant stress and energy signalling.
Nature 448, 938–942.
Bar-Peled, L., Schweitzer, L.D., Zoncu, R., and Sabatini, D.M. (2012). Ragula-
tor is a GEF for the rag GTPases that signal amino acid levels to mTORC1. Cell
150, 1196–1208.
Bar-Peled, L., Chantranupong, L., Cherniack, A.D., Chen, W.W., Ottina, K.A.,
Grabiner, B.C., Spear, E.D., Carter, S.L., Meyerson, M., and Sabatini, D.M.
(2013). A Tumor suppressor complex with GAP activity for the Rag GTPases
that signal amino acid sufficiency to mTORC1. Science 340, 1100–1106.
, D., Berndt, A., Ohashi, Y., Johnson, C.M., and Williams, R.L. (2016).
Baretic
Tor forms a dimer through an N-terminal helical solenoid with a complex topol-
ogy. Nat. Commun. 7, 11016.
Barnes, K., Ingram, J.C., Porras, O.H., Barros, L.F., Hudson, E.R., Fryer, L.G.,
Foufelle, F., Carling, D., Hardie, D.G., and Baldwin, S.A. (2002). Activation of
GLUT1 by metabolic and osmotic stress: potential involvement of AMP-acti-
vated protein kinase (AMPK). J. Cell Sci. 115, 2433–2442.
Bateman, A. (1997). The structure of a domain common to archaebacteria and
the homocystinuria disease protein. Trends Biochem. Sci. 22, 12–13.
Ben-Sahra, I., and Manning, B.D. (2017). mTORC1 signaling and the metabolic
control of cell growth. Curr. Opin. Cell Biol. 45, 72–82.
Ben-Sahra, I., Howell, J.J., Asara, J.M., and Manning, B.D. (2013). Stimulation
of de novo pyrimidine synthesis by growth signaling through mTOR and S6K1.
Science 339, 1323–1328.
Ben-Sahra, I., Hoxhaj, G., Ricoult, S.J.H., Asara, J.M., and Manning, B.D.
(2016). mTORC1 induces purine synthesis through control of the mitochondrial
tetrahydrofolate cycle. Science 351, 728–733.
Benjamin, D., Colombi, M., Moroni, C., and Hall, M.N. (2011). Rapamycin
passes the torch: a new generation of mTOR inhibitors. Nat. Rev. Drug Discov.
10, 868–880.
Binda, M., Péli-Gulli, M.-P., Bonfils, G., Panchaud, N., Urban, J., Sturgill, T.W.,
Loewith, R., and De Virgilio, C. (2009). The Vam6 GEF controls TORC1 by acti-
vating the EGO complex. Mol. Cell 35, 563–573.
Böhme, U., Otto, T.D., Cotton, J.A., Steinbiss, S., Sanders, M., Oyola, S.O.,
Nicot, A., Gandon, S., Patra, K.P., Herd, C., et al. (2018). Complete avian ma-
laria parasite genomes reveal features associated with lineage-specific evolu-
tion in birds and mammals. Genome Res. 28, 547–560.
Brown, E.J., Albers, M.W., Shin, T.B., Ichikawa, K., Keith, C.T., Lane, W.S., and
Schreiber, S.L. (1994). A mammalian protein targeted by G1-arresting rapamy-
cin-receptor complex. Nature 369, 756–758.
Bultot, L., Guigas, B., Von Wilamowitz-Moellendorff, A., Maisin, L., Vertom-
men, D., Hussain, N., Beullens, M., Guinovart, J.J., Foretz, M., Viollet, B.,
et al. (2012). AMP-activated protein kinase phosphorylates and inactivates
liver glycogen synthase. Biochem. J. 443, 193–203.
Buttgereit, F., and Brand, M.D. (1995). A hierarchy of ATP-consuming pro-
cesses in mammalian cells. Biochem. J. 312, 163–167.
Calabrese, M.F., Rajamohan, F., Harris, M.S., Caspers, N.L., Magyar, R.,
Withka, J.M., Wang, H., Borzilleri, K.A., Sahasrabudhe, P.V., Hoth, L.R.,
et al. (2014). Structural basis for AMPK activation: natural and synthetic ligands
regulate kinase activity from opposite poles by different molecular mecha-
nisms. Structure 22, 1161–1172.
Cantó, C., Gerhart-Hines, Z., Feige, J.N., Lagouge, M., Noriega, L., Milne, J.C.,
Elliott, P.J., Puigserver, P., and Auwerx, J. (2009). AMPK regulates energy
expenditure by modulating NAD+ metabolism and SIRT1 activity. Nature
458, 1056–1060.
Carling, D., Zammit, V.A., and Hardie, D.G. (1987). A common bicyclic protein
kinase cascade inactivates the regulatory enzymes of fatty acid and choles-
terol biosynthesis. FEBS Lett. 223, 217–222.
Celenza, J.L., and Carlson, M. (1986). A yeast gene that is essential for release
from glucose repression encodes a protein kinase. Science 233, 1175–1180.
486 Cell Metabolism 31, March 3, 2020
Cell Metabolism
Review
Chantranupong, L., Wolfson, R.L., Orozco, J.M., Saxton, R.A., Scaria, S.M.,
Bar-Peled, L., Spooner, E., Isasa, M., Gygi, S.P., and Sabatini, D.M. (2014).
The Sestrins interact with GATOR2 to negatively regulate the amino-acid-
sensing pathway upstream of mTORC1. Cell Rep. 9, 1–8.
Chantranupong, L., Scaria, S.M., Saxton, R.A., Gygi, M.P., Shen, K., Wyant,
G.A., Wang, T., Harper, J.W., Gygi, S.P., and Sabatini, D.M. (2016). The
CASTOR Proteins Are Arginine Sensors for the mTORC1 Pathway. Cell 165,
153–164.
Chao, L.H., and Avruch, J. (2019). Cryo-EM insight into the structure of MTOR
complex 1 and its interactions with Rheb and substrates. F1000Res. 8, F1000.
Chen, L., Jiao, Z.H., Zheng, L.S., Zhang, Y.Y., Xie, S.T., Wang, Z.X., and Wu,
J.W. (2009). Structural insight into the autoinhibition mechanism of AMP-acti-
vated protein kinase. Nature 459, 1146–1149.
Chen, L., Xin, F.J., Wang, J., Hu, J., Zhang, Y.Y., Wan, S., Cao, L.S., Lu, C., Li,
P., Yan, S.F., et al. (2013). Conserved regulatory elements in AMPK. Nature
498, E8–E10.
Chen, X., Liu, M., Tian, Y., Li, J., Qi, Y., Zhao, D., Wu, Z., Huang, M., Wong,
C.C.L., Wang, H.-W., et al. (2018). Cryo-EM structure of human mTOR com-
plex 2. Cell Research 28, 518–528.
Chia, I.V., and Costantini, F. (2005). Mouse axin and axin2/conductin proteins
are functionally equivalent in vivo. Mol. Cell. Biol. 25, 4371–4376.
Chiu, M.I., Katz, H., and Berlin, V. (1994). RAPT1, a mammalian homolog of
yeast Tor, interacts with the FKBP12/rapamycin complex. Proc. Natl. Acad.
Sci. USA 91, 12574–12578.
Cokorinos, E.C., Delmore, J., Reyes, A.R., Albuquerque, B., Kjobsted, R., Jor-
gensen, N.O., Tran, J.L., Jatkar, A., Cialdea, K., Esquejo, R.M., et al. (2017).
Activation of skeletal muscle AMPK promotes glucose disposal and glucose
lowering in non-human primates and mice. Cell Metab. 25, 1147–1159.e1110.
Collodet, C., Foretz, M., Deak, M., Bultot, L., Metairon, S., Viollet, B., Lefebvre,
G., Raymond, F., Parisi, A., Civiletto, G., et al. (2019). AMPK promotes induc-
tion of the tumor suppressor FLCN through activation of TFEB independently
of mTOR. FASEB J. 33, 12374–12391.
Cool, B., Zinker, B., Chiou, W., Kifle, L., Cao, N., Perham, M., Dickinson, R.,
Adler, A., Gagne, G., Iyengar, R., et al. (2006). Identification and characteriza-
tion of a small molecule AMPK activator that treats key components of type 2
diabetes and the metabolic syndrome. Cell Metab. 3, 403–416.
Dagon, Y., Hur, E., Zheng, B., Wellenstein, K., Cantley, L.C., and Kahn, B.B.
(2012). p70S6 kinase phosphorylates AMPK on serine 491 to mediate leptin’s
effect on food intake. Cell Metab. 16, 104–112.
Davies, S.P., Helps, N.R., Cohen, P.T., and Hardie, D.G. (1995). 50 -AMP in-
hibits dephosphorylation, as well as promoting phosphorylation, of the AMP-
activated protein kinase. Studies using bacterially expressed human protein
phosphatase-2C alpha and native bovine protein phosphatase-2AC. FEBS
Lett. 377, 421–425.
de Duve, C. (2005). The lysosome turns fifty. Nat. Cell Biol. 7, 847–849.
Demetriades, C., Doumpas, N., and Teleman, A.A. (2014). Regulation of
TORC1 in response to amino acid starvation via lysosomal recruitment of
TSC2. Cell 156, 786–799.
Demetriades, C., Plescher, M., and Teleman, A.A. (2016). Lysosomal recruit-
ment of TSC2 is a universal response to cellular stress. Nat. Commun.
7, 10662.
DeVit, M.J., and Johnston, M. (1999). The nuclear exportin Msn5 is required for
nuclear export of the Mig1 glucose repressor of Saccharomyces cerevisiae.
Curr. Biol. 9, 1231–1241.
Dibble, C.C., Elis, W., Menon, S., Qin, W., Klekota, J., Asara, J.M., Finan, P.M.,
Kwiatkowski, D.J., Murphy, L.O., and Manning, B.D. (2012). TBC1D7 is a third
subunit of the TSC1-TSC2 complex upstream of mTORC1. Mol. Cell 47,
535–546.
Dite, T.A., Ling, N.X.Y., Scott, J.W., Hoque, A., Galic, S., Parker, B.L., Ngoei,
K.R.W., Langendorf, C.G., O’Brien, M.T., Kundu, M., et al. (2017). The auto-
phagy initiator ULK1 sensitizes AMPK to allosteric drugs. Nat. Commun.
8, 571.
Ducommun, S., Deak, M., Sumpton, D., Ford, R.J., Núñez Galindo, A., Kuss-
mann, M., Viollet, B., Steinberg, G.R., Foretz, M., Dayon, L., et al. (2015). Motif
Cell Metabolism
Review
affinity and mass spectrometry proteomic approach for the discovery of
cellular AMPK targets: identification of mitochondrial fission factor as a new
AMPK substrate. Cell. Signal. 27, 978–988.
Dunlop, E.A., and Tee, A.R. (2013). The kinase triad, AMPK, mTORC1 and
ULK1, maintains energy and nutrient homoeostasis. Biochem. Soc. Trans.
41, 939–943.
Durán, R.V., Oppliger, W., Robitaille, A.M., Heiserich, L., Skendaj, R., Gottlieb,
E., and Hall, M.N. (2012). Glutaminolysis activates Rag-mTORC1 signaling.
Mol. Cell 47, 349–358.
Durán, R.V., MacKenzie, E.D., Boulahbel, H., Frezza, C., Heiserich, L., Tardito,
S., Bussolati, O., Rocha, S., Hall, M.N., and Gottlieb, E. (2013). HIF-indepen-
dent role of prolyl hydroxylases in the cellular response to amino acids. Onco-
gene 32, 4549–4556.
Efeyan, A., Zoncu, R., Chang, S., Gumper, I., Snitkin, H., Wolfson, R.L., Kirak,
O., Sabatini, D.D., and Sabatini, D.M. (2013). Regulation of mTORC1 by the
Rag GTPases is necessary for neonatal autophagy and survival. Nature 493,
679–683.
Egan, D., Kim, J., Shaw, R.J., and Guan, K.L. (2011a). The autophagy initiating
kinase ULK1 is regulated via opposing phosphorylation by AMPK and mTOR.
Autophagy 7, 643–644.
Egan, D.F., Shackelford, D.B., Mihaylova, M.M., Gelino, S., Kohnz, R.A., Mair,
W., Vasquez, D.S., Joshi, A., Gwinn, D.M., Taylor, R., et al. (2011b). Phosphor-
ylation of ULK1 (hATG1) by AMP-activated protein kinase connects energy
sensing to mitophagy. Science 331, 456–461.
Ferrer, A., Caelles, C., Massot, N., and Hegardt, F.G. (1985). Activation of rat
liver cytosolic 3-hydroxy-3-methylglutaryl coenzyme A reductase kinase by
adenosine 50 -monophosphate. Biochem. Biophys. Res. Commun. 132,
497–504.
Fogarty, S., Ross, F.A., Vara Ciruelos, D., Gray, A., Gowans, G.J., and Hardie,
D.G. (2016). AMPK causes cell cycle arrest in LKB1-deficient cells via activa-
tion of CAMKK2. Mol. Cancer Res. 14, 683–695.
Forte, G.M., Davie, E., Lie, S., Franz-Wachtel, M., Ovens, A.J., Wang, T., Oak-
ek, B., Hagan, I.M., and Petersen, J. (2019). Import of extracel-
hill, J.S., Mac
lular ATP in yeast and man modulates AMPK and TORC1 signalling. J. Cell
Sci. 132, jcs223925.
Foster, K.G., Acosta-Jaquez, H.A., Romeo, Y., Ekim, B., Soliman, G.A., Car-
riere, A., Roux, P.P., Ballif, B.A., and Fingar, D.C. (2010). Regulation of
mTOR complex 1 (mTORC1) by raptor Ser863 and multisite phosphorylation.
J. Biol. Chem. 285, 80–94.
Fu, X., Wan, S., Lyu, Y.L., Liu, L.F., and Qi, H. (2008). Etoposide induces ATM-
dependent mitochondrial biogenesis through AMPK activation. PLoS One
3, e2009.
Fullerton, M.D., Galic, S., Marcinko, K., Sikkema, S., Pulinilkunnil, T., Chen,
Z.P., O’Neill, H.M., Ford, R.J., Palanivel, R., O’Brien, M., et al. (2013). Single
phosphorylation sites in Acc1 and Acc2 regulate lipid homeostasis and the in-
sulin-sensitizing effects of metformin. Nat. Med. 19, 1649–1654.
Gaubitz, C., Oliveira, T.M., Prouteau, M., Leitner, A., Karuppasamy, M., Kon-
stantinidou, G., Rispal, D., Eltschinger, S., Robinson, G.C., Thore, S., et al.
(2015). Molecular Basis of the Rapamycin Insensitivity of Target Of Rapamycin
Complex 2. Mol. Cell 58, 977–988.
González, A., and Hall, M.N. (2017). Nutrient sensing and TOR signaling in
yeast and mammals. EMBO J. 36, 397–408.
Göransson, O., McBride, A., Hawley, S.A., Ross, F.A., Shpiro, N., Foretz, M.,
Viollet, B., Hardie, D.G., and Sakamoto, K. (2007). Mechanism of action of
A-769662, a valuable tool for activation of AMP-activated protein kinase.
J. Biol. Chem. 282, 32549–32560.
Gouw, M., Michael, S., Sámano-Sánchez, H., Kumar, M., Zeke, A., Lang, B.,
Bely, B., Chemes, L.B., Davey, N.E., Deng, Z., et al. (2018). The eukaryotic
linear motif resource - 2018 update. Nucleic Acids Res. 46 (D1), D428–D434.
Gowans, G.J., Hawley, S.A., Ross, F.A., and Hardie, D.G. (2013). AMP is a true
physiological regulator of AMP-activated protein kinase by both allosteric acti-
vation and enhancing net phosphorylation. Cell Metab. 18, 556–566.
Greer, E.L., Oskoui, P.R., Banko, M.R., Maniar, J.M., Gygi, M.P., Gygi, S.P.,
and Brunet, A. (2007). The energy sensor AMP-activated protein kinase
directly regulates the mammalian FOXO3 transcription factor. J. Biol. Chem.
282, 30107–30119.
Gu, X., Orozco, J.M., Saxton, R.A., Condon, K.J., Liu, G.Y., Krawczyk, P.A.,
Scaria, S.M., Harper, J.W., Gygi, S.P., and Sabatini, D.M. (2017a). SAMTOR
is an S-adenosylmethionine sensor for the mTORC1 pathway. Science 358,
813–818.
Gu, X., Yan, Y., Novick, S.J., Kovach, A., Goswami, D., Ke, J., Tan, M.H.E.,
Wang, L., Li, X., de Waal, P.W., et al. (2017b). Deconvoluting AMP-activated
protein kinase (AMPK) adenine nucleotide binding and sensing. J. Biol.
Chem. 292, 12653–12666.
Guri, Y., and Hall, M.N. (2016). mTOR Signaling Confers Resistance to Tar-
geted Cancer Drugs. Trends Cancer 2, 688–697.
Gwinn, D.M., Shackelford, D.B., Egan, D.F., Mihaylova, M.M., Mery, A., Vas-
quez, D.S., Turk, B.E., and Shaw, R.J. (2008). AMPK phosphorylation of raptor
mediates a metabolic checkpoint. Mol. Cell 30, 214–226.
Hall, M.N. (2013). Talks about TORCs: recent advancesin target of rapamycin
signalling. On mTOR nomenclature. Biochem. Soc. Trans. 41, 887–888.
Hara, K., Maruki, Y., Long, X., Yoshino, K., Oshiro, N., Hidayat, S., Tokunaga,
C., Avruch, J., and Yonezawa, K. (2002). Raptor, a binding partner of target of
rapamycin (TOR), mediates TOR action. Cell 110, 177–189.
Hardie, D.G., Carling, D., and Sim, A.T.R. (1989). The AMP-activated protein
kinase - a multisubstrate regulator of lipid metabolism. Trends Biochem. Sci.
14, 20–23.
Hardie, D.G., Schaffer, B.E., and Brunet, A. (2016). AMPK: an energy-sensing
pathway with multiple inputs and outputs. Trends Cell Biol. 26, 190–201.
Hawley, S.A., Selbert, M.A., Goldstein, E.G., Edelman, A.M., Carling, D., and
Hardie, D.G. (1995). 50 -AMP activates the AMP-activated protein kinase
cascade, and Ca2+/calmodulin activates the calmodulin-dependent protein
kinase I cascade, via three independent mechanisms. J. Biol. Chem. 270,
27186–27191.
Hawley, S.A., Davison, M., Woods, A., Davies, S.P., Beri, R.K., Carling, D., and
Hardie, D.G. (1996). Characterization of the AMP-activated protein kinase ki-
nase from rat liver and identification of threonine 172 as the major site at which
it phosphorylates AMP-activated protein kinase. J. Biol. Chem. 271,
27879–27887.
€kela
Hawley, S.A., Boudeau, J., Reid, J.L., Mustard, K.J., Udd, L., Ma €, T.P.,
Alessi, D.R., and Hardie, D.G. (2003). Complexes between the LKB1 tumor
suppressor, STRAD alpha/beta and MO25 alpha/beta are upstream kinases
in the AMP-activated protein kinase cascade. J. Biol. 2, 28.
Hawley, S.A., Pan, D.A., Mustard, K.J., Ross, L., Bain, J., Edelman, A.M., Fren-
guelli, B.G., and Hardie, D.G. (2005). Calmodulin-dependent protein kinase ki-
nase-beta is an alternative upstream kinase for AMP-activated protein kinase.
Cell Metab. 2, 9–19.
Hawley, S.A., Fullerton, M.D., Ross, F.A., Schertzer, J.D., Chevtzoff, C.,
Walker, K.J., Peggie, M.W., Zibrova, D., Green, K.A., Mustard, K.J., et al.
(2012). The ancient drug salicylate directly activates AMP-activated protein ki-
nase. Science 336, 918–922.
Hawley, S.A., Ross, F.A., Gowans, G.J., Tibarewal, P., Leslie, N.R., and Hardie,
D.G. (2014). Phosphorylation by Akt within the ST loop of AMPK-a1 down-reg-
ulates its activation in tumour cells. Biochem. J. 459, 275–287.
Heathcote, H.R., Mancini, S.J., Strembitska, A., Jamal, K., Reihill, J.A., Palmer,
T.M., Gould, G.W., and Salt, I.P. (2016). Protein kinase C phosphorylates AMP-
activated protein kinase a1 Ser487. Biochem. J. 473, 4681–4697.
Hedbacker, K., and Carlson, M. (2008). SNF1/AMPK pathways in yeast. Front.
Biosci. 13, 2408–2420.
Heitman, J., Movva, N.R., and Hall, M.N. (1991). Targets for cell cycle arrest by
the immunosuppressant rapamycin in yeast. Science 253, 905–909.
Herrero-Martı́n, G., Høyer-Hansen, M., Garcı́a-Garcı́a, C., Fumarola, C., Far-
€a
kas, T., López-Rivas, A., and Ja €ttela
€, M. (2009). TAK1 activates AMPK-
dependent cytoprotective autophagy in TRAIL-treated epithelial cells. EMBO
J. 28, 677–685.
Hoppe, S., Bierhoff, H., Cado, I., Weber, A., Tiebe, M., Grummt, I., and Voit, R.
(2009). AMP-activated protein kinase adapts rRNA synthesis to cellular energy
supply. Proc. Natl. Acad. Sci. USA 106, 17781–17786.
Cell Metabolism 31, March 3, 2020 487

Horman, S., Vertommen, D., Heath, R., Neumann, D., Mouton, V., Woods, A.,
Schlattner, U., Wallimann, T., Carling, D., Hue, L., and Rider, M.H. (2006). Insu-
lin antagonizes ischemia-induced Thr172 phosphorylation of AMP-activated
protein kinase alpha-subunits in heart via hierarchical phosphorylation of
Ser485/491. J. Biol. Chem. 281, 5335–5340.
Hudson, E.R., Pan, D.A., James, J., Lucocq, J.M., Hawley, S.A., Green, K.A.,
Baba, O., Terashima, T., and Hardie, D.G. (2003). A novel domain in AMP-acti-
vated protein kinase causes glycogen storage bodies similar to those seen in
hereditary cardiac arrhythmias. Curr. Biol. 13, 861–866.
Hughes Hallett, J.E., Luo, X., and Capaldi, A.P. (2015). Snf1/AMPK promotes
the formation of Kog1/Raptor-bodies to increase the activation threshold of
TORC1 in budding yeast. Elife 4, e09181.
Hurley, R.L., Anderson, K.A., Franzone, J.M., Kemp, B.E., Means, A.R., and
Witters, L.A. (2005). The Ca2+/calmodulin-dependent protein kinase kinases
are AMP-activated protein kinase kinases. J. Biol. Chem. 280, 29060–29066.
Hurley, R.L., Barré, L.K., Wood, S.D., Anderson, K.A., Kemp, B.E., Means,
A.R., and Witters, L.A. (2006). Regulation of AMP-activated protein kinase
by multisite phosphorylation in response to agents that elevate cellular
cAMP. J. Biol. Chem. 281, 36662–36672.
Imamura, K., Ogura, T., Kishimoto, A., Kaminishi, M., and Esumi, H. (2001). Cell
cycle regulation via p53 phosphorylation by a 50 -AMP activated protein kinase
activator, 5-aminoimidazole- 4-carboxamide-1-beta-D-ribofuranoside, in a
human hepatocellular carcinoma cell line. Biochem. Biophys. Res. Commun.
287, 562–567.
Imseng, S., Aylett, C.H.S., and Maier, T. (2018). Architecture and activation of
phosphatidylinositol 3-kinase related kinases. Curr. Opin. Struct. Biol. 49,
177–189.
Inoki, K., Li, Y., Zhu, T., Wu, J., and Guan, K.-L. (2002). TSC2 is phosphorylated
and inhibited by Akt and suppresses mTOR signalling. Nat. Cell Biol. 4,
648–657.
Inoki, K., Zhu, T., and Guan, K.L. (2003). TSC2 mediates cellular energy
response to control cell growth and survival. Cell 115, 577–590.
Jadeja, R.N., Chu, X., Wood, C., Bartoli, M., and Khurana, S. (2019). M3
Muscarinic receptor activation reduces hepatocyte lipid accumulation via
CaMKKbeta/AMPK pathway. Biochem. Pharmacol. 169, 113613.
€ger, S., Handschin, C., St-Pierre, J., and Spiegelman, B.M. (2007). AMP-
Ja
activated protein kinase (AMPK) action in skeletal muscle via direct phosphor-
ylation of PGC-1alpha. Proc. Natl. Acad. Sci. USA 104, 12017–12022.
Jaleel, M., McBride, A., Lizcano, J.M., Deak, M., Toth, R., Morrice, N.A., and
Alessi, D.R. (2005). Identification of the sucrose non-fermenting related kinase
SNRK, as a novel LKB1 substrate. FEBS Lett. 579, 1417–1423.
Jewell, J.L., Kim, Y.C., Russell, R.C., Yu, F.X., Park, H.W., Plouffe, S.W., Ta-
gliabracci, V.S., and Guan, K.L. (2015). Metabolism. Differential regulation of
mTORC1 by leucine and glutamine. Science 347, 194–198.
Jewell, J.L., Fu, V., Hong, A.W., Yu, F.-X., Meng, D., Melick, C.H., Wang, H.,
Lam, W.M., Yuan, H.-X., Taylor, S.S., and Guan, K.L. (2019). GPCR signaling
inhibits mTORC1 via PKA phosphorylation of Raptor. Elife 8, e43038.
Ji, H., Ramsey, M.R., Hayes, D.N., Fan, C., McNamara, K., Kozlowski, P., Tor-
rice, C., Wu, M.C., Shimamura, T., Perera, S.A., et al. (2007). LKB1 modulates
lung cancer differentiation and metastasis. Nature 448, 807–810.
Johanns, M., Pyr Dit Ruys, S., Houddane, A., Vertommen, D., Herinckx, G.,
Hue, L., Proud, C.G., and Rider, M.H. (2017). Direct and indirect activation of
eukaryotic elongation factor 2 kinase by AMP-activated protein kinase. Cell.
Signal. 36, 212–221.
Jørgensen, S.B., Nielsen, J.N., Birk, J.B., Olsen, G.S., Viollet, B., Andreelli, F.,
Schjerling, P., Vaulont, S., Hardie, D.G., Hansen, B.F., et al. (2004). The alpha2-
5’AMP-activated protein kinase is a site 2 glycogen synthase kinase in skeletal
muscle and is responsive to glucose loading. Diabetes 53, 3074–3081.
Jung, J., Genau, H.M., and Behrends, C. (2015). Amino Acid-Dependent
mTORC1 Regulation by the Lysosomal Membrane Protein SLC38A9. Mol.
Cell. Biol. 35, 2479–2494.
Kalender, A., Selvaraj, A., Kim, S.Y., Gulati, P., Brûlé, S., Viollet, B., Kemp, B.E.,
Bardeesy, N., Dennis, P., Schlager, J.J., et al. (2010). Metformin, independent
of AMPK, inhibits mTORC1 in a rag GTPase-dependent manner. Cell Metab.
11, 390–401.
488 Cell Metabolism 31, March 3, 2020
Cell Metabolism
Review
Karuppasamy, M., Kusmider, B., Oliveira, T.M., Gaubitz, C., Prouteau, M.,
Loewith, R., and Schaffitzel, C. (2017). Cryo-EM structure of Saccharomyces
cerevisiae target of rapamycin complex 2. Nat. Commun. 8, 1729.
Kazyken, D., Magnuson, B., Bodur, C., Acosta-Jaquez, H.A., Zhang, D., Tong,
X., Barnes, T.M., Steinl, G.K., Patterson, N.E., Altheim, C.H., et al. (2019).
AMPK directly activates mTORC2 to promote cell survival during acute ener-
getic stress. Sci. Signal. 12, eaav3249.
Keith, C.T., and Schreiber, S.L. (1995). PIK-related kinases: DNA repair,
recombination, and cell cycle checkpoints. Science 270, 50–51.
Kim, J., and Guan, K.-L. (2019). mTOR as a central hub of nutrient signalling
and cell growth. Nat. Cell Biol. 21, 63–71.
Kim, D.-H., Sarbassov, D.D., Ali, S.M., King, J.E., Latek, R.R., Erdjument-
Bromage, H., Tempst, P., and Sabatini, D.M. (2002). mTOR interacts with
raptor to form a nutrient-sensitive complex that signals to the cell growth ma-
chinery. Cell 110, 163–175.
Kim, J., Kundu, M., Viollet, B., and Guan, K.L. (2011). AMPK and mTOR regu-
late autophagy through direct phosphorylation of Ulk1. Nat. Cell Biol. 13,
132–141.
Kim, J., Kim, Y.C., Fang, C., Russell, R.C., Kim, J.H., Fan, W., Liu, R., Zhong,
Q., and Guan, K.L. (2013). Differential regulation of distinct Vps34 complexes
by AMPK in nutrient stress and autophagy. Cell 152, 290–303.
Kim, J.S., Ro, S.-H., Kim, M., Park, H.-W., Semple, I.A., Park, H., Cho, U.-S.,
Wang, W., Guan, K.-L., Karin, M., and Lee, J.H. (2015). Sestrin2 inhibits
mTORC1 through modulation of GATOR complexes. Sci. Rep. 5, 9502.
Kogan, K., Spear, E.D., Kaiser, C.A., and Fass, D. (2010). Structural conserva-
tion of components in the amino acid sensing branch of the TOR pathway in
yeast and mammals. J. Mol. Biol. 402, 388–398.
Kunz, J., Henriquez, R., Schneider, U., Deuter-Reinhard, M., Movva, N.R., and
Hall, M.N. (1993). Target of rapamycin in yeast, TOR2, is an essential phospha-
tidylinositol kinase homolog required for G1 progression. Cell 73, 585–596.
Kurth-Kraczek, E.J., Hirshman, M.F., Goodyear, L.J., and Winder, W.W.
(1999). 50 AMP-activated protein kinase activation causes GLUT4 transloca-
tion in skeletal muscle. Diabetes 48, 1667–1671.
Lane, N. (2006). Power, sex and suicide: mitochondria and the meaning of life
(Oxford University Press).
Lane, N., and Martin, W. (2010). The energetics of genome complexity. Nature
467, 929–934.
Langendorf, C.G., and Kemp, B.E. (2015). Choreography of AMPK activation.
Cell Res. 25, 5–6.
Lantier, L., Fentz, J., Mounier, R., Leclerc, J., Treebak, J.T., Pehmøller, C.,
Sanz, N., Sakakibara, I., Saint-Amand, E., Rimbaud, S., et al. (2014). AMPK
controls exercise endurance, mitochondrial oxidative capacity, and skeletal
muscle integrity. FASEB J. 28, 3211–3224.
Lee, J.H., Cho, U.-S., and Karin, M. (2016). Sestrin regulation of TORC1: Is
Sestrin a leucine sensor? Sci. Signal. 9, re5.
Levine, T.P., Daniels, R.D., Wong, L.H., Gatta, A.T., Gerondopoulos, A., and
Barr, F.A. (2013). Discovery of new Longin and Roadblock domains that
form platforms for small GTPases in Ragulator and TRAPP-II. Small GTPases
4, 62–69.
Li, X., Wang, L., Zhou, X.E., Ke, J., de Waal, P.W., Gu, X., Tan, M.H., Wang, D.,
Wu, D., Xu, H.E., and Melcher, K. (2015). Structural basis of AMPK regulation
by adenine nucleotides and glycogen. Cell Res. 25, 50–66.
Li, M., Zhang, C.S., Zong, Y., Feng, J.W., Ma, T., Hu, M., Lin, Z., Li, X., Xie, C.,
Wu, Y., et al. (2019a). Transient Receptor Potential V Channels Are Essential
for Glucose Sensing by Aldolase and AMPK. Cell Metab. 30, 508–524.e12.
Li, S., Lavagnino, Z., Lemacon, D., Kong, L., Ustione, A., Ng, X., Zhang, Y.,
Wang, Y., Zheng, B., Piwnica-Worms, H., et al. (2019b). Ca(2+)-stimulated
AMPK-dependent phosphorylation of Exo1 protects stressed replication forks
from aberrant resection. Mol. Cell 74, 1123–1137.e1126.
Ling, N.X.Y., Kaczmarek, A., Hoque, A., Davie, E., Ngoei, K.R.W., Morrison,
K.R., Smiles, W.J., Forte, G.M., Wang, T., Lie, S., et al. (2020). Nat. Metab.
mTORC1 directly inhibits AMPK to promote cell proliferation under nutrient
stress 2, 41–49.
Cell Metabolism
Review
Liu, D., Bordicchia, M., Zhang, C., Fang, H., Wei, W., Li, J.-L., Guilherme, A.,
Guntur, K., Czech, M.P., and Collins, S. (2016). Activation of mTORC1 is
essential for b-adrenergic stimulation of adipose browning. J. Clin. Invest.
126, 1704–1716.
Lizcano, J.M., Göransson, O., Toth, R., Deak, M., Morrice, N.A., Boudeau, J.,
Hawley, S.A., Udd, L., Ma €kela
€, T.P., Hardie, D.G., and Alessi, D.R. (2004).
LKB1 is a master kinase that activates 13 kinases of the AMPK subfamily,
including MARK/PAR-1. EMBO J. 23, 833–843.
Loewith, R., Jacinto, E., Wullschleger, S., Lorberg, A., Crespo, J.L., Bonenfant,
D., Oppliger, W., Jenoe, P., and Hall, M.N. (2002). Two TOR complexes, only
one of which is rapamycin sensitive, have distinct roles in cell growth control.
Mol. Cell 10, 457–468.
Loh, K., Tam, S., Murray-Segal, L., Huynh, K., Meikle, P.J., Scott, J.W., van
Denderen, B., Chen, Z., Steel, R., LeBlond, N.D., et al. (2018). Inhibition of
Adenosine Monophosphate-activated Protein Kinase-3-Hydroxy-3-Methyl-
glutaryl Coenzyme A Reductase signaling leads to hypercholesterolemia and
promotes hepatic steatosis and insulin resistance. Hepatol. Commun.
3, 84–98.
Long, X., Lin, Y., Ortiz-Vega, S., Yonezawa, K., and Avruch, J. (2005). Rheb
binds and regulates the mTOR kinase. Curr. Biol. 15, 702–713.
Lumbreras, V., Alba, M.M., Kleinow, T., Koncz, C., and Pagès, M. (2001).
Domain fusion between SNF1-related kinase subunits during plant evolution.
EMBO Rep. 2, 55–60.
Mackintosh, R.W., Davies, S.P., Clarke, P.R., Weekes, J., Gillespie, J.G., Gibb,
B.J., and Hardie, D.G. (1992). Evidence for a protein kinase cascade in higher
plants. 3-Hydroxy-3-methylglutaryl-CoA reductase kinase. Eur. J. Biochem.
209, 923–931.
Marsin, A.S., Bertrand, L., Rider, M.H., Deprez, J., Beauloye, C., Vincent, M.F.,
Van den Berghe, G., Carling, D., and Hue, L. (2000). Phosphorylation and acti-
vation of heart PFK-2 by AMPK has a role in the stimulation of glycolysis during
ischaemia. Curr. Biol. 10, 1247–1255.
Marsin, A.S., Bouzin, C., Bertrand, L., and Hue, L. (2002). The stimulation of
glycolysis by hypoxia in activated monocytes is mediated by AMP-activated
protein kinase and inducible 6-phosphofructo-2-kinase. J. Biol. Chem. 277,
30778–30783.
Martina, J.A., Chen, Y., Gucek, M., and Puertollano, R. (2012). MTORC1 func-
tions as a transcriptional regulator of autophagy by preventing nuclear trans-
port of TFEB. Autophagy 8, 903–914.
Mayer, F.V., Heath, R., Underwood, E., Sanders, M.J., Carmena, D., McCart-
ney, R.R., Leiper, F.C., Xiao, B., Jing, C., Walker, P.A., et al. (2011). ADP reg-
ulates SNF1, the Saccharomyces cerevisiae homolog of AMP-activated pro-
tein kinase. Cell Metab. 14, 707–714.
McBride, A., Ghilagaber, S., Nikolaev, A., and Hardie, D.G. (2009). The
glycogen-binding domain on the AMPK beta subunit allows the kinase to act
as a glycogen sensor. Cell Metab. 9, 23–34.
Menon, S., Dibble, C.C., Talbott, G., Hoxhaj, G., Valvezan, A.J., Takahashi, H.,
Cantley, L.C., and Manning, B.D. (2014). Spatial control of the TSC complex
integrates insulin and nutrient regulation of mTORC1 at the lysosome. Cell
156, 771–785.
Merlin, J., Evans, B.A., Csikasz, R.I., Bengtsson, T., Summers, R.J., and
Hutchinson, D.S. (2010). The M3-muscarinic acetylcholine receptor stimulates
glucose uptake in L6 skeletal muscle cells by a CaMKK-AMPK-dependent
mechanism. Cell. Signal. 22, 1104–1113.
Miranda-Saavedra, D., Stark, M.J., Packer, J.C., Vivares, C.P., Doerig, C., and
Barton, G.J. (2007). The complement of protein kinases of the microsporidium
Encephalitozoon cuniculi in relation to those of Saccharomyces cerevisiae and
Schizosaccharomyces pombe. BMC Genomics 8, 309.
Mitchelhill, K.I., Stapleton, D., Gao, G., House, C., Michell, B., Katsis, F., Wit-
ters, L.A., and Kemp, B.E. (1994). Mammalian AMP-activated protein kinase
shares structural and functional homology with the catalytic domain of yeast
Snf1 protein kinase. J. Biol. Chem. 269, 2361–2364.
Momcilovic, M., Hong, S.P., and Carlson, M. (2006). Mammalian TAK1 acti-
vates Snf1 protein kinase in yeast and phosphorylates AMP-activated protein
kinase in vitro. J. Biol. Chem. 281, 25336–25343.
Moreau, M., Azzopardi, M., Clement, G., Dobrenel, T., Marchive, C., Renne, C.,
Martin-Magniette, M.L., Taconnat, L., Renou, J.P., Robaglia, C., and Meyer, C.
(2012). Mutations in the Arabidopsis homolog of LST8/GbetaL, a partner of the
target of Rapamycin kinase, impair plant growth, flowering, and metabolic
adaptation to long days. Plant Cell 24, 463–481.
Morris, A.J., and Tolan, D.R. (1993). Site-directed mutagenesis identifies
aspartate 33 as a previously unidentified critical residue in the catalytic mech-
anism of rabbit aldolase A. J. Biol. Chem. 268, 1095–1100.
Myers, R.W., Guan, H.P., Ehrhart, J., Petrov, A., Prahalada, S., Tozzo, E.,
Yang, X., Kurtz, M.M., Trujillo, M., Gonzalez Trotter, D., et al. (2017). Systemic
pan-AMPK activator MK-8722 improves glucose homeostasis but induces
cardiac hypertrophy. Science 357, 507–511.
Nada, S., Hondo, A., Kasai, A., Koike, M., Saito, K., Uchiyama, Y., and Okada,
M. (2009). The novel lipid raft adaptor p18 controls endosome dynamics by
anchoring the MEK-ERK pathway to late endosomes. EMBO J. 28, 477–489.
Ngoei, K.R.W., Langendorf, C.G., Ling, N.X.Y., Hoque, A., Varghese, S., Ca-
merino, M.A., Walker, S.R., Bozikis, Y.E., Dite, T.A., Ovens, A.J., et al.
(2018). Structural determinants for small-molecule activation of skeletal mus-
cle AMPK alpha2beta2gamma1 by the glucose importagog SC4. Cell Chem.
Biol. 25, 728–737.e9.
Nicastro, R., Sardu, A., Panchaud, N., and De Virgilio, C. (2017). The Architec-
ture of the Rag GTPase Signaling Network. Biomolecules 7, 48.
Nojima, H., Tokunaga, C., Eguchi, S., Oshiro, N., Hidayat, S., Yoshino, K.,
Hara, K., Tanaka, N., Avruch, J., and Yonezawa, K. (2003). The mammalian
target of rapamycin (mTOR) partner, raptor, binds the mTOR substrates p70
S6 kinase and 4E-BP1 through their TOR signaling (TOS) motif. J. Biol.
Chem. 278, 15461–15464.
Nukarinen, E., Na€gele, T., Pedrotti, L., Wurzinger, B., Mair, A., Landgraf, R.,
Börnke, F., Hanson, J., Teige, M., Baena-Gonzalez, E., et al. (2016). Quantita-
tive phosphoproteomics reveals the role of the AMPK plant ortholog SnRK1 as
a metabolic master regulator under energy deprivation. Sci. Rep. 6, 31697.
O’Donnell, A.F., and Schmidt, M.C. (2019). AMPK-mediated regulation of
alpha-arrestins and protein trafficking. Int. J. Mol. Sci. 20, E515.
O’Neill, H.M., Maarbjerg, S.J., Crane, J.D., Jeppesen, J., Jørgensen, S.B.,
Schertzer, J.D., Shyroka, O., Kiens, B., van Denderen, B.J., Tarnopolsky,
M.A., et al. (2011). AMP-activated protein kinase (AMPK) beta1beta2 muscle
null mice reveal an essential role for AMPK in maintaining mitochondrial con-
tent and glucose uptake during exercise. Proc. Natl. Acad. Sci. USA 108,
16092–16097.
Oakhill, J.S., Steel, R., Chen, Z.P., Scott, J.W., Ling, N., Tam, S., and Kemp,
B.E. (2011). AMPK is a direct adenylate charge-regulated protein kinase. Sci-
ence 332, 1433–1435.
Orlova, M., Kanter, E., Krakovich, D., and Kuchin, S. (2006). Nitrogen availabil-
ity and TOR regulate the Snf1 protein kinase in Saccharomyces cerevisiae. Eu-
karyot. Cell 5, 1831–1837.
Palm, W., and Thompson, C.B. (2017). Nutrient acquisition strategies of
mammalian cells. Nature 546, 234–242.
Pan, D.A., and Hardie, D.G. (2002). A homologue of AMP-activated protein ki-
nase in Drosophila melanogaster is sensitive to AMP and is activated by ATP
depletion. Biochem. J. 367, 179–186.
Panchaud, N., Péli-Gulli, M.-P., and De Virgilio, C. (2013a). Amino acid depri-
vation inhibits TORC1 through a GTPase-activating protein complex for the
Rag family GTPase Gtr1. Sci. Signal. 6, ra42.
Panchaud, N., Péli-Gulli, M.-P., and De Virgilio, C. (2013b). SEACing the GAP
that nEGOCiates TORC1 activation: evolutionary conservation of Rag GTPase
regulation. Cell Cycle 12, 2948–2952.
Papamichos-Chronakis, M., Gligoris, T., and Tzamarias, D. (2004). The Snf1 ki-
nase controls glucose repression in yeast by modulating interactions between
the Mig1 repressor and the Cyc8-Tup1 co-repressor. EMBO Rep. 5, 368–372.
Parmigiani, A., Nourbakhsh, A., Ding, B., Wang, W., Kim, Y.C., Akopiants, K.,
Guan, K.-L., Karin, M., and Budanov, A.V. (2014). Sestrins inhibit mTORC1 ki-
nase activation through the GATOR complex. Cell Rep. 9, 1281–1291.
Pehmøller, C., Treebak, J.T., Birk, J.B., Chen, S., Mackintosh, C., Hardie, D.G.,
Richter, E.A., and Wojtaszewski, J.F. (2009). Genetic disruption of AMPK
signaling abolishes both contraction- and insulin-stimulated TBC1D1 phos-
phorylation and 14-3-3 binding in mouse skeletal muscle. Am. J. Physiol. En-
docrinol. Metab. 297, E665–E675.
Cell Metabolism 31, March 3, 2020 489

Pelosse, M., Cottet-Rousselle, C., Bidan, C.M., Dupont, A., Gupta, K., Berger,
I., and Schlattner, U. (2019). Synthetic energy sensor AMPfret deciphers ad-
enylate-dependent AMPK activation mechanism. Nat. Commun. 10, 1038.
Peng, M., Yin, N., and Li, M.O. (2017). SZT2 dictates GATOR control of
mTORC1 signalling. Nature 543, 433–437.
Petit, C.S., Roczniak-Ferguson, A., and Ferguson, S.M. (2013). Recruitment of
folliculin to lysosomes supports the amino acid-dependent activation of Rag
GTPases. J. Cell Biol. 202, 1107–1122.
Polekhina, G., Gupta, A., Michell, B.J., van Denderen, B., Murthy, S., Feil, S.C.,
Jennings, I.G., Campbell, D.J., Witters, L.A., Parker, M.W., et al. (2003). AMPK
beta subunit targets metabolic stress sensing to glycogen. Curr. Biol. 13,
867–871.
Polekhina, G., Gupta, A., van Denderen, B.J., Feil, S.C., Kemp, B.E., Stapleton,
D., and Parker, M.W. (2005). Structural basis for glycogen recognition by AMP-
activated protein kinase. Structure 13, 1453–1462.
Powis, K., Zhang, T., Panchaud, N., Wang, R., De Virgilio, C., and Ding, J.
(2015). Crystal structure of the Ego1-Ego2-Ego3 complex and its role in pro-
moting Rag GTPase-dependent TORC1 signaling. Cell Res. 25, 1043–1059.
Prouteau, M., Desfosses, A., Sieben, C., Bourgoint, C., Lydia Mozaffari, N.,
Demurtas, D., Mitra, A.K., Guichard, P., Manley, S., and Loewith, R. (2017).
TORC1 organized in inhibited domains (TOROIDs) regulate TORC1 activity.
Nature 550, 265–269.
Puertollano, R., Ferguson, S.M., Brugarolas, J., and Ballabio, A. (2018). The
complex relationship between TFEB transcription factor phosphorylation
and subcellular localization. EMBO J. 37, e98804.
Qian, X., Li, X., Tan, L., Lee, J.H., Xia, Y., Cai, Q., Zheng, Y., Wang, H., Lorenzi,
P.L., and Lu, Z. (2018). Conversion of PRPS hexamer to monomer by AMPK-
mediated phosphorylation inhibits nucleotide synthesis in response to energy
stress. Cancer Discov. 8, 94–107.
Rebsamen, M., Pochini, L., Stasyk, T., de Araújo, M.E.G., Galluccio, M., Kan-
dasamy, R.K., Snijder, B., Fauster, A., Rudashevskaya, E.L., Bruckner, M.,
et al. (2015). SLC38A9 is a component of the lysosomal amino acid sensing
machinery that controls mTORC1. Nature 519, 477–481.
Reihill, J.A., Ewart, M.A., Hardie, D.G., and Salt, I.P. (2007). AMP-activated
protein kinase mediates VEGF-stimulated endothelial NO production. Bio-
chem. Biophys. Res. Commun. 354, 1084–1088.
Robitaille, A.M., Christen, S., Shimobayashi, M., Cornu, M., Fava, L.L., Moes,
S., Prescianotto-Baschong, C., Sauer, U., Jenoe, P., and Hall, M.N. (2013).
Quantitative phosphoproteomics reveal mTORC1 activates de novo pyrimi-
dine synthesis. Science 339, 1320–1323.
Roczniak-Ferguson, A., Petit, C.S., Froehlich, F., Qian, S., Ky, J., Angarola, B.,
Walther, T.C., and Ferguson, S.M. (2012). The transcription factor TFEB links
mTORC1 signaling to transcriptional control of lysosome homeostasis. Sci.
Signal. 5, ra42.
Rogala, K.B., Gu, X., Kedir, J.F., Abu-Remaileh, M., Bianchi, L.F., Bottino,
A.M.S., Dueholm, R., Niehaus, A., Overwijn, D., Fils, A.P., et al. (2019). Struc-
tural basis for the docking of mTORC1 on the lysosomal surface. Science 366,
468–475.
Ross, F.A., Jensen, T.E., and Hardie, D.G. (2016a). Differential regulation by
AMP and ADP of AMPK complexes containing different g subunit isoforms.
Biochem. J. 473, 189–199.
Ross, F.A., MacKintosh, C., and Hardie, D.G. (2016b). AMP-activated protein
kinase: a cellular energy sensor that comes in 12 flavours. FEBS J. 283,
2987–3001.
Sabatini, D.M., Erdjument-Bromage, H., Lui, M., Tempst, P., and Snyder, S.H.
(1994). RAFT1: a mammalian protein that binds to FKBP12 in a rapamycin-
dependent fashion and is homologous to yeast TORs. Cell 78, 35–43.
Sabers, C.J., Martin, M.M., Brunn, G.J., Williams, J.M., Dumont, F.J., Wieder-
recht, G., and Abraham, R.T. (1995). Isolation of a protein target of the
FKBP12-rapamycin complex in mammalian cells. J. Biol. Chem. 270, 815–822.
Sagan, L. (1967). On the origin of mitosing cells. J. Theor. Biol. 14, 255–274.
Saldivia, M., Barquilla, A., Bart, J.-M., Diaz-González, R., Hall, M.N., and Nav-
arro, M. (2013). Target of rapamycin (TOR) kinase in Trypanosoma brucei: an
extended family. Biochem. Soc. Trans. 41, 934–938.
490 Cell Metabolism 31, March 3, 2020
Cell Metabolism
Review
Sancak, Y., Bar-Peled, L., Zoncu, R., Markhard, A.L., Nada, S., and Sabatini,
D.M. (2010). Ragulator-Rag complex targets mTORC1 to the lysosomal sur-
face and is necessary for its activation by amino acids. Cell 141, 290–303.
Sanchez-Cespedes, M., Parrella, P., Esteller, M., Nomoto, S., Trink, B., Eng-
les, J.M., Westra, W.H., Herman, J.G., and Sidransky, D. (2002). Inactivation
of LKB1/STK11 is a common event in adenocarcinomas of the lung. Cancer
Res. 62, 3659–3662.
Sanders, M.J., Ali, Z.S., Hegarty, B.D., Heath, R., Snowden, M.A., and Carling,
D. (2007). Defining the mechanism of activation of AMP-activated protein ki-
nase by the small molecule A-769662, a member of the thienopyridone family.
J. Biol. Chem. 282, 32539–32548.
Sanli, T., Rashid, A., Liu, C., Harding, S., Bristow, R.G., Cutz, J.C., Singh, G.,
Wright, J., and Tsakiridis, T. (2010). Ionizing radiation activates AMP-activated
kinase (AMPK): a target for radiosensitization of human cancer cells. Int. J. Ra-
diat. Oncol. Biol. Phys. 78, 221–229.
Saxton, R.A., and Sabatini, D.M. (2017). mTOR Signaling in Growth, Meta-
bolism, and Disease. Cell 168, 960–976.
Saxton, R.A., Chantranupong, L., Knockenhauer, K.E., Schwartz, T.U., and
Sabatini, D.M. (2016a). Mechanism of arginine sensing by CASTOR1 upstream
of mTORC1. Nature 536, 229–233.
Saxton, R.A., Knockenhauer, K.E., Schwartz, T.U., and Sabatini, D.M. (2016b).
The apo-structure of the leucine sensor Sestrin2 is still elusive. Sci. Signal.
9, ra92.
Saxton, R.A., Knockenhauer, K.E., Wolfson, R.L., Chantranupong, L., Pacold,
M.E., Wang, T., Schwartz, T.U., and Sabatini, D.M. (2016c). Structural basis for
leucine sensing by the Sestrin2-mTORC1 pathway. Science 351, 53–58.
Schaffer, B.E., Levin, R.S., Hertz, N.T., Maures, T.J., Schoof, M.L., Hollstein,
P.E., Benayoun, B.A., Banko, M.R., Shaw, R.J., Shokat, K.M., and Brunet, A.
(2015). Identification of AMPK phosphorylation sites reveals a network of pro-
teins involved in cell invasion and facilitates large-scale substrate prediction.
Cell Metab. 22, 907–921.
Schalm, S.S., and Blenis, J. (2002). Identification of a conserved motif required
for mTOR signaling. Curr. Biol. 12, 632–639.
Schalm, S.S., Fingar, D.C., Sabatini, D.M., and Blenis, J. (2003). TOS motif-
mediated raptor binding regulates 4E-BP1 multisite phosphorylation and func-
tion. Curr. Biol. 13, 797–806.
Scott, J.W., Hawley, S.A., Green, K.A., Anis, M., Stewart, G., Scullion, G.A.,
Norman, D.G., and Hardie, D.G. (2004). CBS domains form energy-sensing
modules whose binding of adenosine ligands is disrupted by disease muta-
tions. J. Clin. Invest. 113, 274–284.
Scott, J.W., Ling, N., Issa, S.M., Dite, T.A., O’Brien, M.T., Chen, Z.P., Galic, S.,
Langendorf, C.G., Steinberg, G.R., Kemp, B.E., and Oakhill, J.S. (2014). Small
molecule drug A-769662 and AMP synergistically activate naive AMPK inde-
pendent of upstream kinase signaling. Chem. Biol. 21, 619–627.
Sekiguchi, T., Kamada, Y., Furuno, N., Funakoshi, M., and Kobayashi, H.
(2014). Amino acid residues required for Gtr1p-Gtr2p complex formation and
its interactions with the Ego1p-Ego3p complex and TORC1 components in
yeast. Genes Cells 19, 449–463.
Settembre, C., Zoncu, R., Medina, D.L., Vetrini, F., Erdin, S., Erdin, S., Huynh,
T., Ferron, M., Karsenty, G., Vellard, M.C., et al. (2012). A lysosome-to-nucleus
signalling mechanism senses and regulates the lysosome via mTOR and
TFEB. EMBO J. 31, 1095–1108.
Shaw, R.J., Kosmatka, M., Bardeesy, N., Hurley, R.L., Witters, L.A., DePinho,
R.A., and Cantley, L.C. (2004). The tumor suppressor LKB1 kinase directly ac-
tivates AMP-activated kinase and regulates apoptosis in response to energy
stress. Proc. Natl. Acad. Sci. USA 101, 3329–3335.
Shen, K., and Sabatini, D.M. (2018). Ragulator and SLC38A9 activate the Rag
GTPases through noncanonical GEF mechanisms. Proc. Natl. Acad. Sci. USA
115, 9545–9550.
Shen, K., Huang, R.K., Brignole, E.J., Condon, K.J., Valenstein, M.L., Chantra-
nupong, L., Bomaliyamu, A., Choe, A., Hong, C., Yu, Z., and Sabatini, D.M.
(2018). Architecture of the human GATOR1 and GATOR1-Rag GTPases com-
plexes. Nature 556, 64–69.
Cell Metabolism
Review
Shen, K., Valenstein, M.L., Gu, X., and Sabatini, D.M. (2019). Arg-78 of Nprl2
catalyzes GATOR1-stimulated GTP hydrolysis by the Rag GTPases. J. Biol.
Chem. 294, 2970–2975.
Shertz, C.A., Bastidas, R.J., Li, W., Heitman, J., and Cardenas, M.E. (2010).
Conservation, duplication, and loss of the Tor signaling pathway in the fungal
kingdom. BMC Genomics 11, 510.
Shimobayashi, M., and Hall, M.N. (2014). Making new contacts: the mTOR
network in metabolism and signalling crosstalk. Nat. Rev. Mol. Cell Biol. 15,
155–162.
Shin, H.J., Kim, H., Oh, S., Lee, J.G., Kee, M., Ko, H.J., Kweon, M.N., Won,
K.J., and Baek, S.H. (2016). AMPK-SKP2-CARM1 signalling cascade in tran-
scriptional regulation of autophagy. Nature 534, 553–557.
Smith, F.C., Davies, S.P., Wilson, W.A., Carling, D., and Hardie, D.G. (1999).
The SNF1 kinase complex from Saccharomyces cerevisiae phosphorylates
the transcriptional repressor protein Mig1p in vitro at four sites within or near
regulatory domain 1. FEBS Lett. 453, 219–223.
Son, S.M., Park, S.J., Lee, H., Siddiqi, F., Lee, J.E., Menzies, F.M., and Ru-
binsztein, D.C. (2019). Leucine signals to mTORC1 via Its metabolite acetyl-
Coenzyme A. Cell Metab. 29, 192–201.e197.
Stahmann, N., Woods, A., Carling, D., and Heller, R. (2006). Thrombin acti-
vates AMP-activated protein kinase in endothelial cells via a pathway involving
Ca2+/calmodulin-dependent protein kinase kinase beta. Mol. Cell. Biol. 26,
5933–5945.
Stahmann, N., Woods, A., Spengler, K., Heslegrave, A., Bauer, R., Krause, S.,
Viollet, B., Carling, D., and Heller, R. (2010). Activation of AMP-activated pro-
tein kinase by vascular endothelial growth factor mediates endothelial angio-
genesis independently of nitric-oxide synthase. J. Biol. Chem. 285,
10638–10652.
Stuttfeld, E., Aylett, C.H., Imseng, S., Boehringer, D., Scaiola, A., Sauer, E.,
Hall, M.N., Maier, T., and Ban, N. (2018). Architecture of the human mTORC2
core complex. Elife 7, e33101.
Sugden, C., Crawford, R.M., Halford, N.G., and Hardie, D.G. (1999a). Regula-
tion of spinach SNF1-related (SnRK1) kinases by protein kinases and phos-
phatases is associated with phosphorylation of the T loop and is regulated
by 50 -AMP. Plant J. 19, 433–439.
Sugden, C., Donaghy, P.G., Halford, N.G., and Hardie, D.G. (1999b). Two
SNF1-related protein kinases from spinach leaf phosphorylate and inactivate
3-hydroxy-3-methylglutaryl-coenzyme A reductase, nitrate reductase, and su-
crose phosphate synthase in vitro. Plant Physiol. 120, 257–274.
Suzuki, T., Bridges, D., Nakada, D., Skiniotis, G., Morrison, S.J., Lin, J.D., Salt-
iel, A.R., and Inoki, K. (2013). Inhibition of AMPK catabolic action by GSK3.
Mol. Cell 50, 407–419.
Tatebe, H., and Shiozaki, K. (2017). Evolutionary Conservation of the Compo-
nents in the TOR Signaling Pathways. Biomolecules 7, 77.
Thelander, M., Olsson, T., and Ronne, H. (2004). Snf1-related protein kinase 1
is needed for growth in a normal day-night light cycle. EMBO J. 23, 1900–1910.
Thornton, C., Sardini, A., and Carling, D. (2008). Muscarinic receptor activation
of AMP-activated protein kinase inhibits orexigenic neuropeptide mRNA
expression. J. Biol. Chem. 283, 17116–17122.
Toyama, E.Q., Herzig, S., Courchet, J., Lewis, T.L., Jr., Losón, O.C., Hellberg,
K., Young, N.P., Chen, H., Polleux, F., Chan, D.C., and Shaw, R.J. (2016).
Metabolism. AMP-activated protein kinase mediates mitochondrial fission in
response to energy stress. Science 351, 275–281.
Treitel, M.A., Kuchin, S., and Carlson, M. (1998). Snf1 protein kinase regulates
phosphorylation of the Mig1 repressor in Saccharomyces cerevisiae. Mol. Cell.
Biol. 18, 6273–6280.
Tsun, Z.-Y., Bar-Peled, L., Chantranupong, L., Zoncu, R., Wang, T., Kim, C.,
Spooner, E., and Sabatini, D.M. (2013). The folliculin tumor suppressor is a
GAP for the RagC/D GTPases that signal amino acid levels to mTORC1.
Mol. Cell 52, 495–505.
van Dam, T.J.P., Zwartkruis, F.J.T., Bos, J.L., and Snel, B. (2011). Evolution of
the TOR pathway. J. Mol. Evol. 73, 209–220.
Vara-Ciruelos, D., Dandapani, M., Gray, A., Egbani, E.O., Evans, A.M., and
Hardie, D.G. (2018). Genotoxic damage activates the AMPK-alpha1 isoform
in the nucleus via Ca2+/CaMKK2 signaling to enhance tumor cell survival.
Mol. Cancer Res. 16, 345–357.
Vara-Ciruelos, D., Dandapani, M., Russell, F.M., Grzes, K.M., Atrih, A., Foretz,
M., Viollet, B., Lamont, D.J., Cantrell, D.A., and Hardie, D.G. (2019). Phenfor-
min, but not metformin, delays development of T cell acute lymphoblastic leu-
kemia/lymphoma via cell-autonomous AMPK activation. Cell Rep. 27, 690–
698.e694.
Vincent, O., Townley, R., Kuchin, S., and Carlson, M. (2001). Subcellular local-
ization of the Snf1 kinase is regulated by specific beta subunits and a novel
glucose signaling mechanism. Genes Dev. 15, 1104–1114.
Wang, L., Lawrence, J.C., Jr., Sturgill, T.W., and Harris, T.E. (2009). Mamma-
lian target of rapamycin complex 1 (mTORC1) activity is associated with phos-
phorylation of raptor by mTOR. J. Biol. Chem. 284, 14693–14697.
Wang, B., Zhao, L., Fish, M., Logan, C.Y., and Nusse, R. (2015a). Self-renew-
ing diploid Axin2(+) cells fuel homeostatic renewal of the liver. Nature 524,
180–185.
Wang, S., Tsun, Z.-Y.Z.Y., Wolfson, R.L., Shen, K., Wyant, G.A., Plovanich,
M.E., Yuan, E.D., Jones, T.D., Chantranupong, L., Comb, W., et al. (2015b).
Metabolism. Lysosomal amino acid transporter SLC38A9 signals arginine suf-
ficiency to mTORC1. Science 347, 188–194.
Wilson, W.A., Hawley, S.A., and Hardie, D.G. (1996). Glucose repression/dere-
pression in budding yeast: SNF1 protein kinase is activated by phosphoryla-
tion under derepressing conditions, and this correlates with a high AMP:ATP
ratio. Curr. Biol. 6, 1426–1434.
Winder, W.W., and Hardie, D.G. (1996). Inactivation of acetyl-CoA carboxylase
and activation of AMP-activated protein kinase in muscle during exercise. Am.
J. Physiol. 270, E299–E304.
Winder, W.W., Holmes, B.F., Rubink, D.S., Jensen, E.B., Chen, M., and Hol-
loszy, J.O. (2000). Activation of AMP-activated protein kinase increases mito-
chondrial enzymes in skeletal muscle. J. Appl. Physiol. 88, 2219–2226.
Wolfson, R.L., and Sabatini, D.M. (2017). The Dawn of the Age of Amino Acid
Sensors for the mTORC1 Pathway. Cell Metab. 26, 301–309.
Wolfson, R.L., Chantranupong, L., Saxton, R.A., Shen, K., Scaria, S.M.,
Cantor, J.R., and Sabatini, D.M. (2016). Sestrin2 is a leucine sensor for the
mTORC1 pathway. Science 351, 43–48.
Wolfson, R.L., Chantranupong, L., Wyant, G.A., Gu, X., Orozco, J.M., Shen, K.,
Condon, K.J., Petri, S., Kedir, J., Scaria, S.M., et al. (2017). KICSTOR recruits
GATOR1 to the lysosome and is necessary for nutrients to regulate mTORC1.
Nature 543, 438–442.
Woods, A., Munday, M.R., Scott, J., Yang, X., Carlson, M., and Carling, D.
(1994). Yeast SNF1 is functionally related to mammalian AMP-activated pro-
tein kinase and regulates acetyl-CoA carboxylase in vivo. J. Biol. Chem.
269, 19509–19515.
Woods, A., Johnstone, S.R., Dickerson, K., Leiper, F.C., Fryer, L.G., Neumann,
D., Schlattner, U., Wallimann, T., Carlson, M., and Carling, D. (2003). LKB1 is
the upstream kinase in the AMP-activated protein kinase cascade. Curr. Biol.
13, 2004–2008.
Woods, A., Dickerson, K., Heath, R., Hong, S.P., Momcilovic, M., Johnstone,
S.R., Carlson, M., and Carling, D. (2005). Ca2+/calmodulin-dependent protein
kinase kinase-beta acts upstream of AMP-activated protein kinase in mamma-
lian cells. Cell Metab. 2, 21–33.
Wu, N., Zheng, B., Shaywitz, A., Dagon, Y., Tower, C., Bellinger, G., Shen,
C.H., Wen, J., Asara, J., McGraw, T.E., et al. (2013). AMPK-dependent degra-
dation of TXNIP upon energy stress leads to enhanced glucose uptake via
GLUT1. Mol. Cell 49, 1167–1175.
Wyant, G.A., Abu-Remaileh, M., Wolfson, R.L., Chen, W.W., Freinkman, E.,
Danai, L.V., Vander Heiden, M.G., and Sabatini, D.M. (2017). mTORC1 Acti-
vator SLC38A9 Is Required to Efflux Essential Amino Acids from Lysosomes
and Use Protein as a Nutrient. Cell 171, 642–654.e12.
Xia, J., Wang, R., Zhang, T., and Ding, J. (2016). Structural insight into the argi-
nine-binding specificity of CASTOR1 in amino acid-dependent mTORC1
signaling. Cell Discov. 2, 16035.
Xiao, B., Sanders, M.J., Underwood, E., Heath, R., Mayer, F.V., Carmena, D.,
Jing, C., Walker, P.A., Eccleston, J.F., Haire, L.F., et al. (2011). Structure of
mammalian AMPK and its regulation by ADP. Nature 472, 230–233.
Cell Metabolism 31, March 3, 2020 491

Xiao, B., Sanders, M.J., Carmena, D., Bright, N.J., Haire, L.F., Underwood, E.,
Patel, B.R., Heath, R.B., Walker, P.A., Hallen, S., et al. (2013). Structural basis
of AMPK regulation by small molecule activators. Nat. Commun. 4, 3017.
Xin, F.J., Wang, J., Zhao, R.Q., Wang, Z.X., and Wu, J.W. (2013). Coordinated
regulation of AMPK activity by multiple elements in the a-subunit. Cell Res. 23,
1237–1240.
Xue, R.Q., Sun, L., Yu, X.J., Li, D.L., and Zang, W.J. (2016). Vagal nerve stim-
ulation improves mitochondrial dynamics via an M3 receptor/CaMKKbeta/
AMPK pathway in isoproterenol-induced myocardial ischaemia. J. Cell. Mol.
Med. 21, 58–71.
Yan, Y., Zhou, X.E., Novick, S.J., Shaw, S.J., Li, Y., Brunzelle, J.S., Hitoshi, Y.,
Griffin, P.R., Xu, H.E., and Melcher, K. (2019). Structures of AMP-activated
protein kinase bound to novel pharmacological activators in phosphorylated,
non-phosphorylated, and nucleotide-free states. J. Biol. Chem. 294, 953–967.
Yang, Y., Atasoy, D., Su, H.H., and Sternson, S.M. (2011). Hunger states
switch a flip-flop memory circuit via a synaptic AMPK-dependent positive
feedback loop. Cell 146, 992–1003.
Yang, H., Rudge, D.G., Koos, J.D., Vaidialingam, B., Yang, H.J., and Pavletich,
N.P. (2013). mTOR kinase structure, mechanism and regulation. Nature 497,
217–223.
Yang, H., Wang, J., Liu, M., Chen, X., Huang, M., Tan, D., Dong, M.-Q., Wong,
C.C.L., Wang, J., Xu, Y., and Wang, H.W. (2016). 4.4 Å Resolution Cryo-EM
structure of human mTOR Complex 1. Protein Cell 7, 878–887.
Yang, H., Jiang, X., Li, B., Yang, H.J., Miller, M., Yang, A., Dhar, A., and Pav-
letich, N.P. (2017). Mechanisms of mTORC1 activation by RHEB and inhibition
by PRAS40. Nature 552, 368–373.
Yeh, L.A., Lee, K.H., and Kim, K.H. (1980). Regulation of rat liver acetyl-CoA
carboxylase. Regulation of phosphorylation and inactivation of acetyl-CoA
carboxylase by the adenylate energy charge. J. Biol. Chem. 255, 2308–2314.
Yoon, M.S., Sun, Y., Arauz, E., Jiang, Y., and Chen, J. (2011). Phosphatidic
acid activates mammalian target of rapamycin complex 1 (mTORC1) kinase
by displacing FK506 binding protein 38 (FKBP38) and exerting an allosteric ef-
fect. J. Biol. Chem. 286, 29568–29574.
Yoon, M.S., Son, K., Arauz, E., Han, J.M., Kim, S., and Chen, J. (2016). Leucyl-
tRNA synthetase activates Vps34 in amino acid-sensing mTORC1 signaling.
Cell Rep. 16, 1510–1517.
492 Cell Metabolism 31, March 3, 2020
Cell Metabolism
Review
Yuan, H.-X., Wang, Z., Yu, F.-X., Li, F., Russell, R.C., Jewell, J.L., and Guan,
K.-L. (2015). NLK phosphorylates Raptor to mediate stress-induced mTORC1
inhibition. Genes Dev. 29, 2362–2376.
Zeng, L., Fagotto, F., Zhang, T., Hsu, W., Vasicek, T.J., Perry, W.L., 3rd, Lee,
J.J., Tilghman, S.M., Gumbiner, B.M., and Costantini, F. (1997). The mouse
Fused locus encodes Axin, an inhibitor of the Wnt signaling pathway that reg-
ulates embryonic axis formation. Cell 90, 181–192.
Zeqiraj, E., Filippi, B.M., Deak, M., Alessi, D.R., and van Aalten, D.M. (2009).
Structure of the LKB1-STRAD-MO25 complex reveals an allosteric mecha-
nism of kinase activation. Science 326, 1707–1711.
Zhang, T., Péli-Gulli, M.-P., Yang, H., De Virgilio, C., and Ding, J. (2012). Ego3
functions as a homodimer to mediate the interaction between Gtr1-Gtr2 and
Ego1 in the ego complex to activate TORC1. Structure 20, 2151–2160.
Zhang, Y.L., Guo, H., Zhang, C.S., Lin, S.Y., Yin, Z., Peng, Y., Luo, H., Shi, Y.,
Lian, G., Zhang, C., et al. (2013). AMP as a low-energy charge signal autono-
mously initiates assembly of AXIN-AMPK-LKB1 complex for AMPK activation.
Cell Metab. 18, 546–555.
Zhang, C.S., Jiang, B., Li, M., Zhu, M., Peng, Y., Zhang, Y.L., Wu, Y.Q., Li, T.Y.,
Liang, Y., Lu, Z., et al. (2014). The lysosomal v-ATPase-Ragulator complex is a
common activator for AMPK and mTORC1, acting as a switch between catab-
olism and anabolism. Cell Metab. 20, 526–540.
Zhang, C.S., Hawley, S.A., Zong, Y., Li, M., Wang, Z., Gray, A., Ma, T., Cui, J.,
Feng, J.W., Zhu, M., et al. (2017). Fructose-1,6-bisphosphate and aldolase
mediate glucose sensing by AMPK. Nature 548, 112–116.
Zhao, R.Q. (2019). Expression, purification and characterization of the plant
Snf1-related protein kinase 1 from Escherichia coli. Protein Expr. Purif.
162, 24–31.
Zoncu, R., Bar-Peled, L., Efeyan, A., Wang, S., Sancak, Y., and Sabatini, D.M.
(2011). mTORC1 senses lysosomal amino acids through an inside-out mech-
anism that requires the vacuolar H(+)-ATPase. Science 334, 678–683.
Zong, H., Ren, J.M., Young, L.H., Pypaert, M., Mu, J., Birnbaum, M.J., and
Shulman, G.I. (2002). AMP kinase is required for mitochondrial biogenesis in
skeletal muscle in response to chronic energy deprivation. Proc. Natl. Acad.
Sci. USA 99, 15983–15987.
Zong, Y., Zhang, C.S., Li, M., Wang, W., Wang, Z., Hawley, S.A., Ma, T., Feng,
J.W., Tian, X., Qi, Q., et al. (2019). Hierarchical activation of compartmentalized
pools of AMPK depends on severity of nutrient or energy stress. Cell Res. 29,
460–473.